Cecil Essentials Of Medicine - 8th Edition

Excel in your medicine rotation, residency, or practice Cecil Medicine, 23rd Edition Expert Consult Premium Edition— Advanced Online Features and Print Goldman & Ausiello ISBN: 978-1-4160-4478-9

Netter’s Internal Medicine, 2nd Edition Book & Online Access Runge & Greganti ISBN: 978-1-4160-4997-5

Instant Work-ups: A Clinical Guide to Medicine O’Connell ISBN: 978-1-4160-5296-8

Medical Secrets, 5th Edition Harward ISBN: 978-0-3230-6398-2

Textbook of Physical Diagnosis, 6th Edition

Physical Diagnosis Secrets, 2nd Edition

With Student Consult Online Access Book and DVD Swartz ISBN: 978-1-4160-6203-5

With Student Consult Online Access Mangione ISBN: 978-0-3230-3467-8

Practical Guide to the Care of the Medical Patient, 8th Edition

Review 2 Rounds: Visual Review and Clinical Reference

Expert Consult: Online and Print Ferri ISBN: 978-0-3230-7158-1

Medicine: A CompetencyBased Companion

With Student Consult Online Access Tunkel & Israel ISBN: 978-1-4160-5351-4

Gallardo & Berry ISBN: 978-1-4377-0169-2

Images from the Wards: Diagnosis and Treatment Studdiford et al. ISBN: 978-1-4160-6383-4

Order your copies today! 1-800-545-2522 www.elsevierhealth.com

ANDREOLI and CARPENTER’S Cecil

Essentials of

MEDICINE 8th EDITION

This page intentionally left blank

ANDREOLI and CARPENTER’S Cecil

Essentials of

MEDICINE 8th EDITION Editor-in-Chief THOMAS E. ANDREOLI, MD, MACP, FRCP (Edin.), FRCP (London), ScD (hon.), Docteur (hon.), MD (hon.)†

Distinguished Professor Nolan Chair Emeritus Department of Internal Medicine Department of Physiology and Biophysics University of Arkansas College of Medicine Little Rock, Arkansas (†Deceased)

Editors Ivor J. Benjamin, MD, FACC, FAHA Professor of Medicine Adjunct Professor of Biochemistry Christi T. Smith Endowed Chair for Cardiovascular Research Director, Center for Cardiovascular Translational Biomedicine University of Utah School of Medicine Salt Lake City, Utah

Robert C. Griggs, MD, FACP, FAAN Professor of Neurology, Medicine, Pediatrics, and Pathology and Laboratory Medicine University of Rochester School of Medicine and Dentistry Rochester, New York

Edward J. Wing, MD, FACP, FIDSA Dean of Medicine and Biological Sciences The Warren Alpert Medical School of Brown University Providence, Rhode Island

1600 John F. Kennedy Blvd. Ste 1800 Philadelphia, PA 19103-2899 ANDREOLI AND CARPENTER’S CECIL ESSENTIALS OF MEDICINE

ISBN: 978-1-4160-6109-0

International Edition

ISBN: 978-0-8089-2428-9

Copyright © 2010 Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. With respect to any drug or pharmaceutical products identified, readers are advised to check the most current information provided (i) on procedures featured or (ii) by the manufacturer of each product to be administered, to verify the recommended dose or formula, the method and duration of administration, and contraindications. It is the responsibility of practitioners, relying on their own experience and knowledge of their patients, to make diagnoses, to determine dosages and the best treatment for each individual patient, and to take all appropriate safety precautions. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. Previous editions copyrighted 2007, 2004, 2001, 1997, 1993, 1990, 1986 by Saunders, an imprint of Elsevier Inc. Library of Congress Cataloging-in-Publication Data Andreoli and Carpenter’s Cecil essentials of medicine / editor-in-chief, Thomas E. Andreoli; editors, Ivor J. Benjamin, Robert C. Griggs, Edward J. Wing.—8th ed. p. ; cm. Includes bibliographical references and index. ISBN 978-1-4160-6109-0 1. Internal medicine—Textbooks. I. Andreoli, Thomas E., 1935-2009 II. Cecil, Russell L. (Russell La Fayette), 1881-1965. III. Title: Cecil essentials of medicine. IV. Title: Essentials of medicine. [DNLM: 1. Internal Medicine. WB 115 A559 2010] RC46.C42 2010 616—dc22 2009027158 Cover: Hemoglobin subunit: Phantatomix / Photo Researchers, Inc.; False-color (computer graphics) photograph of a resin cast of the human bronchial tree, the network of airways serving both lungs: Alfred Pasieka / Photo Researchers, Inc.; DNA: Dr. A. Lesk, MRC-LMB / Photo Researchers, Inc.; Osteoarthritis of foot, X-ray: DR P. MARAZZI / Photo Researchers, Inc. Acquisitions Editor: James Merritt Managing Editor: Rebecca Gruliow Publishing Services Manager: Linda Van Pelt Project Manager: Sharon Lee Design Direction: Steven Stave Printed in China Last digit is the print number: 9 8 7 6 5 4 3 2 1

Working together to grow libraries in developing countries www.elsevier.com | www.bookaid.org | www.sabre.org

Dedication Thomas E. Andreoli When Dr. Thomas Andreoli died of a cerebral hemorrhage on April 14, 2009, he had nearly completed the editorial oversight of this, the eighth edition of the textbook he co-founded in 1986. In many ways this textbook epitomizes his career as an educator, clinical scientist, and international leader of the medical profession. As an outstanding, life-long investigator in his chosen field of nephrology, Dr. Andreoli served as president of both national and international medical societies and was recognized by honorary doctoral degrees from both European and American universities. As a dedicated bedside clinician and teacher, Dr. Andreoli served as an outstanding chair of medicine, and endowed chairs in his name were established at both the University of Alabama School of Medicine and the University of Arkansas College of Medicine. His superlative teaching was recognized by his receiving, inter alia, the Louis Pasteur Award from the University Louis Pasteur, Mastership of the American College of Physicians, and the Robert H. Williams Distinguished Chair of Medicine Award from the Association of Professors of Medicine. Perhaps Dr. Andreoli’s most distinguished contribution was his lifelong Oslerian devotion to translating medical science from bench to bedside. Despite his major national and international commitments, he continued throughout his career to hold morning resident teaching rounds five times weekly, maintaining a broad knowledge of all aspects of internal medicine and genuinely and gently transmitting that knowledge to two generations of medical students. He was uniquely qualified for, and committed to, imparting his wisdom and skill as a physician, which provided the basis for his serving as a founding editor of Essentials of Medicine, and editor-in-chief of its last three editions. We feel immensely privileged to have been his co-editors and friends, and we dedicate this text to Dr. Thomas Andreoli. Charles C. J. Carpenter, MD, MACP Professor of Medicine Brown Medical School Director, Brown University AIDS Center Providence, Rhode Island

Clementine M. Whitman No tribute to Dr. Andreoli’s accomplishments would be complete without acknowledging the contributions of Clementine Whitman, his personal assistant of 40 years, who moved with him from Alabama to Texas and to Arkansas. Clementine was the hub of Dr. Andreoli’s professional and personal life, meticulously handling every detail. Dr. Andreoli, a man as demanding of others as of himself, was indeed blessed and fortunate to have such a talented, dedicated, loyal, and hard-working person by his side. Sudhir V. Shah, MD, FACP Professor of Medicine Director, Division of Nephrology University of Arkansas College of Medicine Little Rock, Arkansas Chief, Renal Section, Medicine Service John L. McClellan Memorial Veterans Hospital Little Rock, Arkansas

This page intentionally left blank

Andreoli and Carpenter’s Cecil Essentials of Medicine International Advisory Board NAME

DISCIPLINE

COUNTRY

Professor J. S. Bajaj Chief Consultant and Director Department of Diabetes, Endocrine and Metabolic Medicine Batra Hospital and Medical Research Centre New Delhi, India [emailprotected]

Endocrinology

India

Professor Massimo G. Colombo, MD Professor and Chairman Department of Gastroenterology and Endocrinology IRCCS Maggiore Hospital University of Milan Milan, Italy [emailprotected]

Hepatology/Gastroenterology

Italy

Professor Bertrand Fontaine, MD Professor of Neurology Faculté de Médecine Fédération des Maladies du Système Nerveux Groupe Hospitalier Pitié-Salpêtrière Paris, France [emailprotected]

Neurology

France

Professor Arnoldo Guzmán-Sanchez Professor and Chair Department of Obstetrics and Gynecology Hospital Civil de Guadalajara Guadalajara, Jalisco, Mexico [emailprotected]

Women’s Health

Mexico

Professor Kiyoshi Kurokawa, MD President, Science Council of Japan Professor Emeritus, University of Tokyo Roppongi, Minato-ku Tokyo, Japan [emailprotected]

Nephrology

Japan

Professor Umesh G. Lalloo, MD, FCCP Head, Department of Pulmonology and HIV Nelson R. Mandela School of Medicine University of Kwa-Zulu Natal Durban, South Africa [emailprotected]

General Internal Medicine/HIV

South Africa

Professor Pal Magyar Head, Department of Pulmonary Medicine Semmelweis University Budapest, Hungary [emailprotected]

Pulmonary Medicine

Hungary

Professor John Newsom-Davis, MD Emeritus Professor of Clinical Neurology Radcliffe Infirmary Woodstock Road Oxford, United Kingdom [emailprotected]

Neurology

United Kingdom

NAME

DISCIPLINE

COUNTRY

Professor J. N. Pande Professor of Medicine Department of Infectious Diseases Sita Ram Bhartia Institute of Science and Research All India Institute of Medical Sciences New Delhi, India [emailprotected]

Infectious Diseases

India

Dr. Mario Paredes-Espinoza Professor and Chair Department of Internal Medicine Hospital Civil Fray Antonio Alcalde Guadalajara, Jalisco, Mexico [emailprotected]

Men’s Health/General Internal Medicine

Mexico

Professor Nestor Schor, MD, PhD Head Professor of Medicine Nephrology Division UNIFESP-Escola Paulista de Medicina São Paulo, Brazil [emailprotected]

Nephrology

Brazil

Lead Authors and Contributors Section I Introduction to Molecular Medicine Lead Author Ivor J. Benjamin, MD, FACC, FAHA Professor of Medicine Adjunct Professor of Biochemistry Christi T. Smith Endowed Chair in Cardiovascular Research Director, Center for Cardiovascular Translational Biomedicine University of Utah School of Medicine Salt Lake City, Utah [emailprotected]

Section II Evidence-Based Medicine Lead Authors Sara G. Tariq, MD Associate Professor Department of Internal Medicine University of Arkansas College of Medicine Little Rock, Arkansas [emailprotected] Susan S. Beland, MD Associate Professor Department of Internal Medicine University of Arkansas College of Medicine Little Rock, Arkansas [emailprotected]

Section III Cardiovascular Disease Lead Author Ivor J. Benjamin, MD, FACC, FAHA Professor of Medicine Adjunct Professor of Biochemistry Christi T. Smith Endowed Chair in Cardiovascular Research Director, Center for Cardiovascular Translational Biomedicine University of Utah School of Medicine Salt Lake City, Utah [emailprotected] Contributors David Bull, MD Professor of Surgery Director, Thoracic Surgery Residency Program University of Utah School of Medicine Chief of Cardiothoracic Surgery Salt Lake City VA Medical Center Salt Lake City, Utah [emailprotected]

Mohamed H. Hamdan, MD Professor and Associate Division Chief Division of Cardiology University of Utah School of Medicine Section Chief, Arrhythmia University of Utah Healthcare Salt Lake City, Utah [emailprotected] Dean Y. Li, MD, PhD Associate Professor Departments of Medicine and Oncological Science Huntsman Cancer Institute University of Utah School of Medicine Salt Lake City, Utah [emailprotected] Sheldon E. Litwin, MD Amundsen Professor of Internal Medicine/Cardiology Director of Cardiovascular Imaging University of Utah Hospital University of Utah School of Medicine Salt Lake City, Utah [emailprotected] Andrew D. Michaels, MD Associate Professor of Internal Medicine Director, Cardiac Catheterization Laboratory and Interventional Cardiology University of Utah School of Medicine Salt Lake City, Utah [emailprotected] Jack H. Morshedzadeh, MD Instructor Division of Cardiology University of Utah School of Medicine Salt Lake City, Utah [emailprotected] Josef Stehlik, MD Assistant Professor of Internal Medicine Division of Cardiology University of Utah School of Medicine Salt Lake City, Utah [emailprotected] Kevin J. Whitehead, MD Associate Professor of Cardiology University of Utah School of Medicine Salt Lake City, Utah [emailprotected] Ronald G. Victor, MD Associate Director Cedars-Sinai Heart Institute Director, Cedars-Sinai Hypertension Center Los Angeles, California [emailprotected]

Lead Authors and Contributors

Wanpen Vongpatanasin, MD Associate Professor of Internal Medicine-Cardiology The University of Texas Southwestern Medical School Dallas, Texas [emailprotected]

Section IV Pulmonary and Critical Care Medicine Lead Author Sharon I. Rounds, MD Professor of Medicine and of Pathology and Laboratory Medicine Brown Medical School Chief of Pulmonary Critical Care Medicine Providence VA Medical Center Providence, Rhode Island [emailprotected] Contributors Jason M. Aliotta, MD Assistant Professor of Medicine Division of Biology and Medicine Brown University Providence, Rhode Island [emailprotected] Brian Casserly, MD Assistant Professor of Medicine Brown University Providence, Rhode Island [emailprotected] Matthew D. Jankowich, MD Instructor in Medicine Division of Biology and Medicine Brown University Providence, Rhode Island [emailprotected] F. Dennis McCool, MD Chief, Pulmonary Critical Care Medicine Memorial Hospital of Rhode Island Professor of Medicine Alpert Medical School of Brown University Pawtucket, Rhode Island [emailprotected]

Section V Preoperative and Postoperative Care Lead Author Kim A. Eagle, MD Albion Walter Hewlett Professor of Internal Medicine Chief, Clinical Cardiovascular Medicine Director, Cardiovascular Center University of Michigan Medical School Ann Arbor, Michigan [emailprotected]

Contributors Wei C. Lau, MD Clinical Associate Professor Director, Adult Cardiovascular Thoracic Anesthesiology Medical Director, Cardiovascular Center Operating Rooms Department of Anesthesiology University of Michigan Health System Ann Arbor, Michigan [emailprotected]

Section VI Renal Disease Lead Author Raymond C. Harris, MD Ann and Roscoe R. Robinson Professor of Medicine Director, Division of Nephrology Vanderbilt University School of Medicine Nashville, Tennessee [emailprotected] Contributors Thomas E. Andreoli, MD, MACP, FRCP (Edinburgh), FRCP (London), ScD (hon.), Docteur (hon.), MD (hon.), Doctor (hon.) Distinguished Professor of Internal Medicine and of Physiology and Biophysics Nolan Chairman Emeritus of Internal Medicine University of Arkansas College of Medicine Little Rock, Arkansas Amanda W. Basford, MD Kidney Associates, PLLC 6624 Fannin, Suite 1400 Houston, Texas [emailprotected] Kerri L. Cavanaugh, MD Assistant Professor of Medicine Vanderbilt University School of Medicine Nashville, Tennessee [emailprotected] Jamie P. Dwyer, MD Assistant Professor of Medicine, Nephrology and Hypertension Vanderbilt University School of Medicine Nashville, Tennessee [emailprotected] Thomas A. Golper, MD Professor of Medicine/Nephrology Director, Medical Specialties Patient Care Center Vanderbilt University School of Medicine Nashville, Tennessee [emailprotected] Michelle W. Krause, MD, MPH Assistant Professor of Medicine Division of Nephrology Department of Internal Medicine University of Arkansas College of Medicine Little Rock, Arkansas [emailprotected]

Lead Authors and Contributors T. Alp Ikizler, MD Catherine McLaughlin Hakim Professor of Medicine Director, Clinical Research in Nephrology Director, Master of Science in Clinical Investigation Program Medical Director, Vanderbilt Outpatient Dialysis Unit Division of Nephrology Vanderbilt University School of Medicine Nashville, Tennessee [emailprotected] Julia B. Lewis, MD Professor of Medicine Director, Fellowship Training Division of Nephrology Vanderbilt University School of Medicine Nashville, Tennessee [emailprotected] James M. Luther, MD, MSCI Assistant Professor of Medicine Division of Nephrology Vanderbilt University School of Medicine Nashville, Tennessee [emailprotected] James L. Pirkle, MD Nephrology and Hypertension Specialists, P.C. Dalton, Georgia [emailprotected] Didier Portilla, MD Professor of Medicine Division of Nephrology Department of Internal Medicine University of Arkansas College of Medicine Little Rock, Arkansas [emailprotected] Robert L. Safirstein, MD Professor, Executive Vice Chair Department of Internal Medicine University of Arkansas College of Medicine Chief of Medical Services Central Arkansas Veterans Hospital Little Rock, Arkansas [emailprotected] Gerald Schulman, MD Professor of Medicine Division of Nephrology Vanderbilt University School of Medicine Nashville, Tennessee [emailprotected] Sudhir V. Shah, MD Professor and Director Division of Nephrology Department of Internal Medicine University of Arkansas College of Medicine Little Rock, Arkansas [emailprotected]

Roy Zent, MD Associate Professor of Medicine, Cancer Biology, and Cell and Developmental Biology Division of Nephrology Vanderbilt University School of Medicine Nashville, Tennessee [emailprotected]

Section VII Gastrointestinal Disease Lead Author M. Michael Wolfe, MD Professor of Medicine Research Professor of Physiology and Biophysics Boston University School of Medicine Chief, Section of Gastroenterology Boston Medical Center Boston, Massachusetts 02118 [emailprotected] Contributors Wanda P. Blanton, MD Instructor, Department of Medicine Section of Gastroenterology Boston University School of Medicine Boston, Massachusetts [emailprotected] Charles M. Bliss Jr., MD, FACP Assistant Professor of Medicine Section of Gastroenterology Boston University School of Medicine Boston, Massachusetts [emailprotected] Francis A. Farraye, MD, MSc Clinical Director, Section of Gastroenterology Co-Director, Center for Digestive Disorders Professor of Medicine Boston University School of Medicine Boston, Massachusetts [emailprotected] Christopher S. Huang, MD Instructor of Medicine Section of Gastroenterology Boston University School of Medicine Boston, Massachusetts [emailprotected] Brian C. Jacobson, MD, MPH Director of Endoscopic Ultrasonography Associate Director of Endoscopy Services Boston Medical Center and Assistant Professor of Medicine Boston University School of Medicine Boston, Massachusetts [emailprotected]

xii

Lead Authors and Contributors

David R. Lichtenstein, MD, FACG Director of Gastrointestinal Endoscopy Associate Professor of Medicine Boston University School of Medicine Boston, Massachusetts [emailprotected] Robert Lowe, MD Associate Professor of Medicine Educational Director of the Section of Gastroenterology Boston University School of Medicine Boston, Massachusetts [emailprotected] Daniel S. Mishkin, MD, CM Director, The Endoscopy Center of Brookline Instructor of Medicine Boston University School of Medicine Boston, Massachusetts [emailprotected] T. Carlton Moore, MD Assistant Professor in Medicine Section of Gastroenterology Boston University School of Medicne Boston, Massachusetts [emailprotected] Jaime A. Oviedo, MD, FACG Greater Boston Gastroenterology Framingham, Massachusetts [emailprotected] Marcos C. Pedrosa, MD, MPH Chief of Endoscopy VA Boston HealthCare System Brigham and Women’s Hospital Boston, Massachusetts [emailprotected] Elihu M. Schimmel, MD Director VA Advanced Specialty Training Program in Gastroenterology and Hepatology Boston VA Hospital Boston, Massachusetts [emailprotected] Paul C. Schroy, III, MD, MPH Director of Clinical Research Section of Gastroenterology Associate Professor of Medicine Boston University School of Medicine Associate Professor of Epidemiology/Biostatistics Boston University School of Public Health Boston, Massachusetts [emailprotected] Satish K. Singh, MD Assistant Professor of Medicine Boston University School of Medicine Staff Gastroenterologist VA Boston HealthCare System Boston, MA 02118 [emailprotected]

Chi-Chuan Tseng, MD, PhD Associate Professor of Medicine Department of Medicine Boston University School of Medicine Associate Chief Boston Veterans Administration Health Care System Boston, Massachusetts [emailprotected]

Section VIII Diseases of the Liver and Biliary System Lead Author Michael B. Fallon, MD Professor of Medicine Director, Division of Gastroenterology, Hepatology and Nutrition The University of Texas Medical School Houston, Texas [emailprotected] Contributors Miguel R. Arguedas MD, MPH Assistant Professor Division of Gastroenterology University of Alabama School of Medicine MCLM 280 Birmingham, Alabama [emailprotected] Rudolf Garcia-Gallont, MD Head, Department of Surgery Amedesgua Hospital Guatemala City, Guatemala [emailprotected] Rajan Kochar, MD Assistant Professor of Medicine Division of Gastroenterology, Hepatology and Nutrition The University of Texas Medical School Houston, Texas [emailprotected] Brendan M. McGuire, MD, MS Associate Professor Medical Director, Liver Transplantation/Medicine Liver Center, Department of Medicine University of Alabama School of Medicine Birmingham, Alabama [emailprotected] Klaus Mönkemüller, MD Associate Professor Chief, Endoscopy and Outpatient Clinic Division of Gastroenterology, Hepatology and Infectious Diseases Otto-von-Guericke University Magdeburg, Germany [emailprotected]

Lead Authors and Contributors Helmut Neumann, MD Faculty of Medicine Division of Gastroenterology, Hepatology and Infectious Diseases Otto-von-Guericke University Magdeburg, Germany [emailprotected] Aasim M. Sheikh, MD Northwest Georgia Gastroenterology Associates Marietta, Georgia [emailprotected] Shyam Varadarajulu, MD Assistant Professor, Division of Gastroenterology Director, Interventional Endoscopy University of Alabama at Birmingham Birmingham, Alabama [emailprotected]

Section IX Hematologic Disease Lead Author Nancy Berliner, MD Professor of Medicine Harvard Medical School Chief, Division of Hematology Brigham and Women’s Hospital Boston, Massachusetts [emailprotected] Contributors Jill Lacy, MD Associate Professor of Medical Oncology Yale University School of Medicine New Haven, Connecticut [emailprotected] Christine S. Rinder, MD Associate Professor of Anesthesiology and Laboratory Medicine Yale University School of Medicine Department of Anesthesiology Yale-New Haven Hospital New Haven, Connecticut [emailprotected] Henry M. Rinder, MD Professor of Laboratory Medicine and Internal Medicine Director, Clinical Hematology Laboratory Program Director, Clinical Pathology Residency Training Yale University School of Medicine New Haven, Connecticut [emailprotected] Michal G. Rose, MD Associate Professor of Medicine Yale University School of Medicine Chief, Cancer Center VA Connecticut HealthCare System New Haven, Connecticut [emailprotected]

xiii

Stuart E. Seropian, MD Associate Professor of Medicine Yale Cancer Center Lymphoma, Leukemia and Myeloma Program New Haven, Connecticut [emailprotected] Christopher Tormey, MD Instructor, Laboratory Medicine Yale University School of Medicine New Haven, Connecticut [emailprotected] Richard Torres, MD Attending Hematopathologist Yale University School of Medicine New Haven, Connecticut [emailprotected] Eunice S. Wang, MD Research Assistant Professor Leukemia Service, Departments of Medicine and Immunology Staff Physician Leukemia Service Roswell Park Cancer Institute Buffalo, New York [emailprotected]

Section X Oncologic Disease Lead Author Jennifer J. Griggs, MD, MPH Associate Professor Department of Internal Medicine Division of Hematology/Oncology Director, Breast Cancer Survivorship Program University of Michigan Comprehensive Cancer Center University of Michigan Medical School Ann Arbor, Michigan [emailprotected] Contributors Barbara A. Burtness, MD Medical Oncologist Fox Chase Cancer Center Philadelphia, Pennsylvania [emailprotected] Alok A. Khorana, MD, FACP Assistant Professor of Medicine, James P. Wilmot Cancer Center University of Rochester School of Medicine and Dentistry Rochester, New York [emailprotected] Paula M. Lantz, MD Professor and Chair Department of Health Management and Policy Research Professor, Institute for Social Research University of Michigan Health System Ann Arbor, Michigan [emailprotected]

xiv

Lead Authors and Contributors

Robert F. Todd, III, MD, PhD Margaret M. Alkek Distinguished Chair and Professor Department of Medicine Baylor College of Medicine Houston, Texas [emailprotected]

Section XI Metabolic Disease Lead Author Robert J. Smith, MD Director, Division of Endocrinology Director, Hallett Center for Diabetes and Endocrinology Brown University Alpert Medical School Providence, Rhode Island [emailprotected] Contributors David G. Brooks, MD, PhD Medical Director, Global Clinical Development Abraxis Bioscience, LLC Burlington, Massachusetts [emailprotected] Geetha Gopalakrishnan, MD Assistant Professor Division of Biology and Medicine Brown University Alpert Medical School Providence, Rhode Island [emailprotected]

Section XII Endocrine Disease Section Author Glenn D. Braunstein, MD Professor of Medicine UCLA School of Medicine Chair, Department of Medicine Cedars-Sinai Medical Center Los Angeles, California [emailprotected] Contributors Philip S. Barnett, MD, PhD Director Anna and Max Webb and Family Diabetes Outpatient Treatment and Education Center Cedars Sinai Medical Center Professor of Medicine David Geffen School of Medicine University of California, Los Angeles Los Angeles, California [emailprotected] Vivien S. Herman-Bonert, MD Associate Professor of Medicine David Geffen School of Medicine Division of Endocrinology University of California, Los Angeles Attending Physician, Cedars Sinai Medical Center Los Angeles, California [emailprotected]

Osama Hamdy, MD Medical Director Obesity Clinical Program Joslin Diabetes Center Assistant Professor of Medicine Harvard Medical School Boston, Massachusetts [emailprotected]

Theodore C. Friedman, MD, PhD Associate Professor of Medicine UCLA School of Medicine Endocrinology Division Cedars Sinai Medical Center Los Angeles, California [emailprotected]

Michelle P. Warren, MD Wyeth-Ayerst Professor Founder and Medical Director Center for Menopause, Hormonal Disorders and Women’s Health Department of Obstetrics and Gynecology Columbia University College of Physicians and Surgeons New York, New York [emailprotected]

Section XIII Women’s Health

Thomas R. Ziegler, MD Professor of Medicine Atlanta Clinical and Translational Science Institute Emory University School of Medicine Atlanta, Georgia [emailprotected]

Contributors Patricia I. Carney, MD Department of Obstetrics and Gynecology Christiana Care Health Services Newark, Delaware [emailprotected]

Lead Author Pamela A. Charney, MD Assistant Professor of Medicine Weill Cornell Medical College New York, New York [emailprotected]

Lead Authors and Contributors Deborah B. Ehrenthal, MD, FACP Departments of Internal Medicine and Obstetrics and Gynecology Christiana Care Health Services Newark, Delaware Clinical Assistant Professor of Medicine Thomas Jefferson University Philadelphia, Pennsylvania [emailprotected] Renee K. Kottenhahn, MD Department of Pediatrics Christiana Care Health Services Newark, Delaware [emailprotected]

Section XIV Men’s Health Lead Author Joseph A. Smith, Jr., MD Professor and Chair Department of Urologic Surgery Vanderbilt University School of Medicine Nashville, Tennessee [emailprotected] Contributors Douglas F. Milam, MD Associate Professor Department of Urologic Surgery Vanderbilt University School of Medicine Nashville, Tennessee [emailprotected] Johnathan S. Starkman, MD Clinical Instructor Department of Urologic Surgery Vanderbilt University School of Medicine Nashville, Tennessee [emailprotected]

Section XV Diseases of Bone and Bone Mineral Metabolism Lead Author Andrew F. Stewart, MD Professor of Medicine Chief, Division of Endocrinology and Metabolism University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania [emailprotected] Contributors Susan L. Greenspan, MD Professor of Medicine Director, Osteoporosis Prevention and Treatment Center Associate Program Director, General Clinical Research Center University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania [emailprotected]

Steven P. Hodak, MD Clinical Assistant Professor of Medicine Medical Director, Center for Diabetes and Endocrinology University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania [emailprotected] Mara J. Horwitz, MD Assistant Professor of Medicine Division of Endocrionology University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania [emailprotected] Shane O. LeBeau, MD Clinical Assistant Professor of Medicine Center for Diabetes and Endocrinology University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania [emailprotected] G. David Roodman, MD, PhD Professor of Medicine University of Pittsburgh Hillman Cancer Center Pittsburgh, Pennsylvania [emailprotected]

Section XVI Musculoskeletal and Connective Tissue Disease Lead Author Larry W. Moreland, MD Margaret Jane Miller Endowed Professor of Arthritis Research Chief, Division of Rheumatology and Clinical Immunology University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania [emailprotected] Contributors Surabhi Agarwal, MD Medical Resident University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania [emailprotected] Dana P. Ascherman, MD Assistant Professor of Medicine Divison of Rheumatology and Clinical Immunology University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania [emailprotected] Robyn T. Domsic, MD Assistant Professor of Medicine Division of Rheumatology and Clinical Immunology University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania [emailprotected]

xvi

Lead Authors and Contributors

Jennifer Rae Elliott, MD Division of Rheumatology and Clinical Immunology University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania [emailprotected] Amy H. Kao, MD, MPH Assistant Professor of Medicine Division of Rheumatology and Clinical Immunology University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania [emailprotected]

Section XVII Infectious Disease Lead Author Edward J. Wing, MD, FACP, FIDSA Dean of Medicine and Biological Sciences The Warren Alpert Medical School of Brown University Providence, Rhode Island [emailprotected]

Fotios Koumpouras, MD Assistant Professor of Medicine Medical Director, Lupus Center of Excellence Division of Rheumatology and Clinical Immunology University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania [emailprotected]

Contributors Keith B. Armitage, MD Professor of Medicine Vice Chair for Education Department of Medicine Co-Director, Medicine/Pediatrics Residency Director, Internal Medicine Residency Training Program Case Western Reserve University Cleveland, Ohio [emailprotected]

C. Kent Kwoh, MD Professor of Medicine Division of Rheumatology and Clinical Immunology University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania [emailprotected]

Curt G. Beckwith, MD Assistant Professor of Medicine Division of Infectious Diseases Brown Medical School Providence, Rhode Island [emailprotected]

Douglas W. Lienesch, MD Assistant Professor of Medicine Division of Rheumatology and Clinical Immunology University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania [emailprotected]

David A. Bobak, MD Associate Professor Division of Infectious Diseases Case Western Reserve University University Hospitals of Cleveland Cleveland, Ohio [emailprotected]

Kathleen McKinnon-Maksimowicz, DO Assistant Professor of Medicine Division of Rheumatology and Clinical Immunology University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania [emailprotected] Thomas A. Medsger, Jr., MD Gerald P. Rodnan Professor of Medicine Division of Rheumatology and Clinical Immunology Director, Scleroderma Research Program University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania [emailprotected] Niveditha Mohan, MD Assistant Professor of Medicine Division of Rheumatology and Clinical Immunology University of Pittsburgh School of Medicine Pittsburgh, Pennsylvania [emailprotected]

Jessica K. Fairley, MD Division of Infectious Disease and HIV Medicine Case Western Reserve University Division of Infectious Disease University Hospitals Cleveland, Ohio [emailprotected] Scott A. Fulton, MD Assistant Professor Division of Infectious Diseases Case Western Reserve University University Hospitals of Cleveland Cleveland, Ohio [emailprotected] Corrilynn O. Hileman, MD Internal Medicine Infectious Disease Case Western Reserve University Cleveland, Ohio [emailprotected]

Lead Authors and Contributors Christoph Lange, MD, PhD Medical Clinic Borstel Research Center Borstel, Germany [emailprotected], [emailprotected] Michael M. Lederman, MD Scott R. Inkley Professor of Medicine Case Western Reserve University Co-Director, CWRU/University Hospitals of Cleveland Center for AIDS Research Cleveland, Ohio [emailprotected] Tracy L. Lemonovich, MD Instructor Division of Infectious Diseases and HIV Medicine Case Western Reserve University Cleveland, Ohio [emailprotected] Michelle V. Lisagaris, MD Assistant Professor Department of Medicine University Hospitals of Cleveland Cleveland, Ohio [emailprotected] Amy J. Ray, MD Clinical Instructor and Division Chief University Hospitals Richmond Medical Center Infectious Diseases Division University Hospitals, School of Medicine Case Western Reserve University Cleveland, Ohio [emailprotected] Benigno Rodriguez, MD Assistant Professor Department of Medicine Case Western Reserve University Cleveland, Ohio [emailprotected] Robert A. Salata, MD Division Chief Division of Infectious Diseases and HIV Medicine Case Western Reserve University Cleveland, Ohio [emailprotected] Richard R. Watkins, MD, MS Division of Infectious Diseases Akron General Medical Center Akron, Ohio

xvii

Section XVIII Bioterrorism Lead Author Robert W. Bradsher, Jr., MD Richard V. Ebert Professor of Internal Medicine Vice-Chair for Education, Department of Internal Medicine Director, Division of Infectious Diseases University of Arkansas College of Medicine Little Rock, Arkansas [emailprotected]

Section XIX Neurologic Disease Lead Author Robert C. Griggs, MD, FACP, FAAN Professor of Neurology, Medicine, Pathology and Laboratory Medicine, and Pediatrics University of Rochester School of Medicine and Dentistry Rochester, New York [emailprotected] Contributors Michel J. Berg, MD Associate Professor of Neurology and Medical Director, Strong Epilepsy Center University of Rochester School of Medicine and Dentistry Rochester, New York [emailprotected] Emma Ciafaloni, MD Associate Professor Department of Neurology (SMD) University of Rochester School of Medicine and Dentistry Rochester, New York [emailprotected] Timothy J. Counihan, MD, MRCPI Department of Neurology Galway University Hospital Galway, Ireland [emailprotected] William P. Cheshire Jr., MD Professor of Neurology Mayo Clinic Jacksonville, Florida [emailprotected] Emily C. de los Reyes, MD Associate Professor of Clinical Pediatrics and Neurology Nationwide Children’s Hospital The Ohio State University Columbus, Ohio [emailprotected]

xviii

Lead Authors and Contributors

Jennifer J. Griggs, MD, MPH Associate Professor Department of Internal Medicine Division of Hematology/Oncology Director, Breast Cancer Survivorship Program University of Michigan Comprehensive Cancer Center University of Michigan Medical School Ann Arbor, Michigan [emailprotected] Carlayne E. Jackson, MD Professor of Neurology University of Texas Medical School San Antonio, Texas [emailprotected] Kevin A. Kerber, MD Assistant Professor Department of Neurology Director, Dizziness Clinic University of Michigan Medical School Ann Arbor, Michigan [emailprotected] Lynn C. Liu, MD Chief, Strong Sleep Disorders Center Department of Neurology University of Rochester School of Medicine and Dentistry Rochester, New York [emailprotected] Geoffrey S.F. Ling, MD, PhD Defense Advanced Research Projects Agency Defense Sciences Office Arlington, Virginia [emailprotected] Jeffery M. Lyness, MD Professor and Associate Chair for Education Department of Psychiatry Director of Curriculum, Office of Curriculum and Assessment University of Rochester School of Medicine and Dentistry Rochester, New York [emailprotected] Deborah Joanne Lynn, MD Associate Professor The Ohio State University Department of Neurology Director, Department of Neurology Medical Student Education Staff Neurologist The Ohio State University Medical Center and The Arthur James Cancer Hospital and Research Institute Co-director, Ohio State University Multiple Sclerosis Center Columbus, Ohio [emailprotected] Frederick J. Marshall, MD Associate Professor Department of Neurology (SMD) University of Rochester Rochester, New York [emailprotected]

Allan McCarthy, MD, MRCPI Department of Neurology Galway University Hospital Galway, Ireland [emailprotected] Sinéad M. Murphy, BA, MB, BCh, MRCPI Department of Neurology Galway University Hospital Galway, Ireland [emailprotected] Avindra Nath, MD Professor of Neurology Johns Hopkins University Baltimore, Maryland [emailprotected] E. Steve Roach, MD Vice Chair for Clinical Affairs Department of Pediatrics Director, Division of Pediatric Neurology Professor of Child Neurology Nationwide Children’s Hospital The Ohio State University Columbus, Ohio [emailprotected] Lisa R. Rogers, DO Director, Medical Neuro-Oncology University Hospitals—Case Medical Center and Professor of Neurology Department of Neurology Case Western University School of Medicine Cleveland, Ohio [emailprotected] Roger P. Simon, MD Chair and Director R.S. Dow Neurobiology Laboratories Legacy Research Hospital and Adjunct Professor Neurology, Physiology and Pharmacology Oregon Health and Science University Portland, Oregon [emailprotected]

Section XX The Aging Patient Lead Author Harvey J. Cohen, MD Walter Kempner Professor and Chair of Medicine Director, Center for the Study of Aging and Human Development Duke University School of Medicine Durham, North Carolina [emailprotected]

Lead Authors and Contributors

xix

Contributor Mitchell T. Heflin, MD, MHS Assistant Professor of Medicine and Geriatrics Center for the Study of Aging and Human Development Duke University School of Medicine Durham, NC Durham, North Carolina [emailprotected]

Section XXII Alcohol and Substance Abuse

Section XXI Palliative Care

Richard A. Lange, MD Professor and Executive Vice-Chair Department of Medicine University of Texas Medical School San Antonio, Texas [emailprotected]

Lead Authors Timothy E. Quill, MD Professor of Medicine, Psychiatry and Medical Humanities Director, Palliative Care Program University of Rochester School of Medicine Rochester, New York [emailprotected] Robert G. Holloway, MD, MPH Professor, Department of Neurology Professor, Department of Community and Preventive Medicine (SMD) Rochester, New York [emailprotected]

Lead Authors L. David Hillis, MD Dan Parman Distinguished Professor Chair, Department of Internal Medicine University of Texas Medical School San Antonio, Texas [emailprotected]

This page intentionally left blank

Preface This is the eighth edition of Andreoli and Carpenter’s Cecil Essentials of Medicine. Essentials VIII, like its predecessors, is intended to be comprehensive but concise. Essentials VIII therefore provides an exacting and thoroughly updated treatise on internal medicine, without excessive length, for students of medicine at all levels of their careers. We welcome with enthusiasm a new editor, Edward J. Wing, MD, Frank L. Day Professor of Biology, and Dean of Medicine and Biological Sciences at Brown University Warren Alpert Medical School. Essentials VIII has three cardinal components. First, at the beginning of each section—kidney, for example—we provide a brief but rigorous summary of the fundamental biology of the kidney and/or the cardinal signs and symptoms of diseases of the kidney. The same format has been used in all the sections of the book. Second, the main body of each section contains a detailed, but again, concise description of the diseases of the various organ systems, together with their pathophysiology and their treatment. Finally, Essentials relies heavily on the Internet. Essentials VIII is published entirely on a Web site on the Internet. In the online version of Essentials VIII, we provide a substantial amount of supplemental material, indicated in the hard copy text by boldface symbols (for example, Web Fig. 1) and

denoted by an arrow icon shown in the margin of this page. This icon is present throughout the hard copy of the book as well as in the Internet version and directs the reader to a series of illustrations, tables, or videos in the Internet version of Essentials. This material is clearly crucial to understanding modern medicine, but we hope that, in this manner, the supplemental material will enrich Essentials VIII without having enlarged the book significantly. As in prior editions, we make abundant use of 4-color illustrations. And as in prior editions, each section has been reviewed by one or another of the editors, and finally by the editor-in-chief. We thank James T. Merritt, Senior Acquisitions Editor, Medical Education, of Elsevier, Inc., and especially Rebecca Gruliow, Managing Editor for Global Medicine, Elsevier, Inc. Both Jim Merritt and Rebecca Gruliow contributed heartily to the preparation of this eighth edition of Essentials. Lastly, we thank our very able secretarial staff, Ms. Clementine M. Whitman (Little Rock); Ms. Barbara S. Bottone (Providence); Ms. Shirley E. Thomas (Rochester); Ms. Jennifer F. Schroff (Salt Lake City); and Ms. Jean M. Drinan, and Ms. Catarina A. Santos (Providence). The Editors

xxi

This page intentionally left blank

Contents SECTION I

Introduction to Molecular Medicine

1 Molecular Basis of Human Disease

Ivor J. Benjamin

2 Evidence-Based Medicine, Quality of Life, and the Cost of Medicine Sara G. Tariq and Susan S. Beland

15 16

SECTION III

Cardiovascular Disease

3 Structure and Function of the Normal Heart and Blood Vessels Jack Morshedzadeh, Dean Y. Li and Ivor J. Benjamin

4 Evaluation of the Patient with Cardiovascular Disease

Sheldon E. Litwin and Ivor J. Benjamin

192

16 Evaluating Lung Structure and Function

198

17 Obstructive Lung Diseases

213

18 Interstitial Lung Diseases

225

19 Pulmonary Vascular Disease

241

20 Disorders of Respiratory Control

245

21 Disorders of the Pleura, Mediastinum, and Chest Wall

248

22 Infectious Diseases of the Lung

254

23 Essentials In Critical Care Medicine

259

24 Neoplastic Disorders of the Lung

266

Brian Casserly and Sharon Rounds

F. Dennis McCool

SECTION II

Evidence-Based Medicine

15 General Approach to Patients with Respiratory Disorders

21 22

Matthew D. Jankowich Jason M. Aliotta and Matthew D. Jankowich Sharon Rounds

Sharon Rounds and Matthew D. Jankowich

F. Dennis McCool

Brian Casserly and Sharon Rounds

5 Diagnostic Tests and Procedures in the Patient with Cardiovascular Disease

6 Heart Failure and Cardiomyopathy

7 Congenital Heart Disease

8 Acquired Valvular Heart Disease

Preoperative and Postoperative Care

273

9 Coronary Heart Disease

25 Preoperative and Postoperative Care

274

Sheldon E. Litwin

Sheldon E. Litwin and Ivor J. Benjamin Kevin J. Whitehead Sheldon E. Litwin

Andrew D. Michaels

Brian Casserly and Sharon Rounds Matthew D. Jankowich and Jason M. Aliotta

SECTION V

Wel C. Lau and Kim A. Eagle

10 Cardiac Arrhythmias

118

11 Pericardial and Myocardial Disease

145

Renal Disease

12 Other Cardiac Topics

156

26 Elements of Renal Structure and Function

286

27 Approach to the Patient with Renal Disease

298

28 Fluid and Electrolyte Disorders

305

29 Glomerular Diseases

323

Mohamed H. Hamdan

Josef Stehlik and Ivor J. Benjamin

David A. Bull and Ivor J. Benjamin

13 Vascular Diseases and Hypertension Wanpen Vongpatanasin and Ronald G. Victor

Robert L. Safirstein

165

Michelle W. Krause, Thomas A. Golper, Raymond C. Harris and Sudhir V. Shah

SECTION IV

Pulmonary and Critical Care Medicine

187

14 The Lung in Health and Disease

188

Sharon Rounds and Matthew D. Jankowich

SECTION VI

Thomas E. Andreoli and Robert L. Safirstein

Jamie P. Dwyer and Julia B. Lewis

xxiii

xxiv

Contents

30 Major Nonglomerular Disorders

333

44 Fulminant Hepatic Failure

476

31 Vascular Disorders of the Kidney

345

45 Cirrhosis of the Liver and its Complications

478

46 Disorders of the Gallbladder and Biliary Tract

488

James L. Pirkle, Amanda W. Basford and Roy Zent James M. Luther and Gerald Schulman

32 Acute Kidney Injury

359

33 Chronic Renal Failure

369

Didier Portilla and Sudhir V. Shah Kerri Cavanaugh and T. Alp Ikizler

Rajan Kochar, Miguel R. Arguedas and Michael B. Fallon

Shyam Varadarajulu, Rudolf Garcia-Gallont and Michael B. Fallon

SECTION VII

Gastrointestinal Disease

Brendan M. McGuire and Michael B. Fallon

381

SECTION IX

34 Common Clinical Manifestations of Gastrointestinal Disease

382

Hematologic Disease

495

A. Abdominal Pain

382

496

B. Gastrointestinal Hemorrhage

385

47 Hematopoiesis and Hematopoietic Failure

48 Clonal Disorders of the Hematopoietic Stem Cell

507

49 Disorders of Red Blood Cells

520

50 Clinical Disorders of Neutrophils

533

51 Disorders of Lymphocytes

539

52 Normal Hemostasis

555

53 Disorders of Hemostasis: Bleeding

564

54 Disorders of Hemostasis: Thrombosis

580

Charles M. Bliss, Jr. and M. Michael Wolfe T. Carlton Moore, Chi-Chuan Tseng and M. Michael Wolfe

C. Malabsorption

389

D. Diarrhea

396

35 Endoscopic and Imaging Procedures

401

36 Esophageal Disorders

408

Marcos C. Pedrosa and Elihu M. Schimmel Satish K. Singh Brian C. Jacobson and Daniel S. Mishkin Robert C. Lowe and M. Michael Wolfe

37 Diseases of the Stomach and Duodenum

Wanda P. Blanton, Jaime A. Oviedo and M. Michael Wolfe

38 Inflammatory Bowel Disease Christopher S. Huang and Francis A. Farraye

414

Michal G. Rose and Nancy Berliner Michal G. Rose and Nancy Berliner Jill Lacy and Stuart Seropian

Christopher A. Tormey and Henry M. Rinder

430

Richard Torres and Henry M. Rinder

439

40 Diseases of the Pancreas

445

David R. Lichtenstein

Eunice S. Wang and Nancy Berliner

Christine S. Rinder and Henry M. Rinder

39 Neoplasms of the Gastrointestinal Tract Paul C. Schroy III

Eunice S. Wang and Nancy Berliner

SECTION VIII

SECTION X

Oncologic Disease

593

55 Cancer Biology and Etiologic Factors

594

Diseases of the Liver and Biliary System

455

Alok A. Khorana and Barbara A. Burtness

41 Laboratory Tests in Liver Disease

456

56 Cancer Epidemiology and Cancer Prevention

598

57 Solid Tumors

603

58 Complications of Cancer and Cancer Treatment

616

Rajan Kochar and Michael B. Fallon

42 Jaundice

Klaus Mönkemüller, Helmut Neumann and Michael B. Fallon

43 Acute and Chronic Hepatitis

Rajan Kochar, Aasim M. Sheikh and Michael B. Fallon

460

Paula M. Lantz and Jennifer J. Griggs Robert F. Todd lll and Jennifer J. Griggs

466

Alok A. Khorana and Jennifer J. Griggs

Contents

59 Principles of Cancer Therapy Alok A. Khorana and Barbara A. Burtness

621

SECTION XI

Metabolic Disease

629

60 Obesity

630

Osama Hamdy and Robert J. Smith

61 Anorexia Nervosa and Bulimia Nervosa Michelle P. Warren

635

62 Malnutrition, Nutritional Assessment, and Nutritional Support in Adult Patients

638

63 Disorders of Lipid Metabolism

643

Thomas R. Ziegler

Geetha Gopalakrishnan and Robert J. Smith

64 Disorders of Metals and Metalloproteins David G. Brooks

651

SECTION XIV

Men’s Health

751

72 Men’s Health Topics

752

A. Benign Prostatic Hyperplasia

752

B. Prostatitis

757

C. Erectile Dysfunction

759

D. Carcinomas of Men

763

E. Benign Scrotal Diseases

768

Jonathan S. Starkman, Douglas F. Milam and Joseph A. Smith, Jr.

SECTION XV

Diseases of Bone and Bone Mineral Metabolism

771

73 Normal Physiology of Bone and Mineral Homeostasis

772

74 Disorders of Serum Minerals

783

75 Metabolic Bone Diseases

795

Andrew F. Stewart

Steven P. Hodak and Andrew F. Stewart

SECTION XII

xxv

Endocrine Disease

659

65 Hypothalamic-Pituitary Axis

660

76 Osteoporosis

802

66 Thyroid Gland

670

77 Paget Disease of Bone

811

67 Adrenal Gland

679

68 Male Reproductive Endocrinology

691

69 Diabetes Mellitus

697

70 Hypoglycemia

721

Vivien S. Herman-Bonert

Vivien S. Herman-Bonert and Theodore C. Friedman Theodore C. Friedman Glenn D. Braunstein

Philip S. Barnett and Glenn D. Braunstein Philip S. Barnett

Shane O. LeBeau and Andrew F. Stewart Susan L. Greenspan Mara J. Horwitz and G. David Roodman

SECTION XVI

Musculoskeletal and Connective Tissue Disease

817

78 Approach to the Patient with Rheumatic Disease

818

79 Rheumatoid Arthritis

823

80 Spondyloarthropathies

829

Niveditha Mohan

Larry W. Moreland

Douglas W. Lienesch

SECTION XIII

Women’s Health

729

81 Systemic Lupus Erythematosus

834

71 Women’s Health Topics

730

82 Antiphospholipid Antibody Syndrome

841

A. Introduction

730

83 Systemic Sclerosis (Scleroderma)

844

B. Preventive Health Recommendations for Women

733

84 Idiopathic Inflammatory Myopathies

850

C. Health Issues Across the Life Course

734

D. Special Topics

745

85 Sjögren Syndrome

855

Deborah Ehrenthal, Patricia Carney, Renee Kottenhahn and Pamela Charney

Jennifer Rae Elliot

Surabhi Agarwal and Amy H. Kao Robyn T. Domsic and Thomas A. Medsger, Jr. Larry W. Moreland

Fotios Koumpouras

xxvi

Contents

86 Systemic Vasculitis

858

87 Crystal Arthropathies

864

88 Osteoarthritis

870

89 Nonarticular Soft Tissue Disorders

873

Kathleen Maksimowicz-Mckinnon Dana P. Ascherman C. Kent Kwoh

Niveditha Mohan

90 Rheumatic Manifestations of Systemic Disorders Fotios Koumpouras

878

883

91 Organisms that Infect Humans

884

Benigno Rodríguez and Michael M. Lederman

890

93 Laboratory Diagnosis of Infectious Diseases

898

94 Antimicrobial Therapy

904

Benigno Rodríguez and Michael M. Lederman Benigno Rodríguez and Michael M. Lederman

95 Fever and Febrile Syndromes

910

96 Bacteremia and Sepsis Syndrome

925

97 Infections of the Nervous System

933

98 Infections of the Head and Neck

945

Tracy L. Lemonovich and Robert A. Salata Richard R. Watkins and Robert A. Salata Scott A. Fulton and Robert A. Salata Christoph Lange and Michael M. Lederman

99 Infections of the Lower Respiratory Tract Christoph Lange and Michael M. Lederman

100 Infections of the Heart and Blood Vessels Benigno Rodríguez and Michael M. Lederman

101 Skin and Soft Tissue Infections Christoph Lange and Michael M. Lederman

102 Intra-Abdominal Abscess and Peritonitis Christoph Lange and Michael M. Lederman

104 Infections Involving Bones and Joints

985

105 Infections of the Urinary Tract

989

106 Health Care−Associated Infections

992

107 Sexually Transmitted Infections

998

108 Human Immunodeficiency Virus Infection and Acquired Immunodeficiency Syndrome

1008

109 Infections in the Immunocompromised Host

1028

110 Infectious Diseases of Travelers: Protozoal and Helminthic Infections

1034

Christoph Lange and Michael M. Lederman

Amy J. Ray, Michelle V. Lisgaris and Robert A. Salata

Infectious Disease

92 Host Defenses Against Infection

979

Christoph Lange and Michael M. Lederman

SECTION XVII

Benigno Rodríguez and Michael M. Lederman

103 Infectious Diarrhea

Corrilynn O. Hileman, Keith B. Armitage and Robert A. Salata

Curt G. Beckwith, Edward J. Wing, Benigno Rodríguez and Michael M. Lederman

Tracy Lemonovich, David A. Bobak and Robert A. Salata

Jessica K. Fairley, Keith B. Armitage and Robert A. Salata

SECTION XVIII

951

Bioterrorism

1043

111 Bioterrorism

1044

Robert W. Bradsher, Jr.

SECTION XIX

Neurologic Disease

1051

112 Neurologic Evaluation of the Patient

1052

113 Disorders of Consciousness

1058

114 Disorders of Sleep

1064

115 Cortical Syndromes

1068

116 Dementia and Memory Disturbances

1072

117 Major Disorders of Mood, Thoughts, and Behavior

1077

Frederick J. Marshall Roger P. Simon

961

969

Lynn Liu

Sinéad M. Murphy and Timothy J. Counihan Frederick J. Marshall

975

Jeffrey M. Lyness

Contents

118 Disorders of Thermal Regulation William P. Cheshire, Jr.

119 Headache, Neck Pain, and Other Painful Disorders Timothy J. Counihan

120 Disorders of Vision and Hearing

Allan McCarthy and Timothy J. Counihan

1083

1086 1096

121 Dizziness and Vertigo

1104

122 Disorders of the Motor System

1108

123 Developmental and Neurocutaneous Disorders

1119

124 Cerebrovascular Disease

1123

Kevin A. Kerber

Frederick J. Marshall

Emily C. de los Reyes and E. Steve Roach Sinéad M. Murphy and Timothy J. Counihan

125 Traumatic Brain Injury and Spinal Cord Injury Geoffrey S. F. Ling

1136

xxvii

130 Neuromuscular Diseases: Disorders of the Motor Neuron and Plexus and Peripheral Nerve Disease

1171

131 Muscle Diseases

1182

132 Neuromuscular Junction Disease

1191

Carlayne E. Jackson Robert C. Griggs Emma Ciafaloni

SECTION XX

The Aging Patient

1195

133 The Aging Patient

1196

Mitchell T. Heflin and Harvey Jay Cohen

SECTION XXI

Palliative Care

1209

134 Palliative Care

1210

Robert G. Holloway and Timothy E. Quill

SECTION XXII

126 Epilepsy

1141

Alcohol and Substance Abuse

1219

127 Central Nervous System Tumors

1154

135 Alcohol and Substance Abuse

1220

128 Infectious Diseases of the Nervous System

1159

Index

1235

129 Demyelinating and Inflammatory Disorders

1165

Michel J. Berg Lisa R. Rogers and Jennifer J. Griggs

Avindra Nath

Deborah Joanne Lynn

Richard A. Lange and L. David Hillis

This page intentionally left blank

Section I Introduction to Molecular Medicine

Molecular Basis of Human Disease –

BENJAMIN

Chapter

Molecular Basis of Human Disease Ivor J. Benjamin

edicine has evolved dramatically during the past century from a healing art in which standards of practice were established on the basis of personal experience, passed on from one practitioner to the next, to a rigorous intellectual discipline steeped in the scientific method. The scientific method, a process that tests the validity of a hypothesis or prediction through experimentation, has led to major advances in the fields of physiology, microbiology, biochemistry, and pharmacology. These advances served as the basis for the diagnostic and therapeutic approaches to illness in common use by physicians through most of the 20th century. Since the 1980s, the understanding of the molecular basis of genetics has expanded dramatically, and advances in this field have identified new and exciting dimensions for defining the basis of conventional genetic diseases (e.g., sickle cell disease) as well as the basis of complex genetic traits (e.g., hypertension). The molecular basis for the interaction between genes and environment has also begun to be defined. Armed with a variety of sensitive and specific molecular techniques, contemporary physicians can now begin not only to understand the molecular underpinning of complex pathobiologic processes but also to identify individuals at risk for common diseases. Understanding modern medicine, therefore, requires an understanding of molecular genetics and the molecular basis of disease. This introductory chapter offers an overview of this complex and rapidly evolving topic and attempts to summarize the principles of molecular medicine that will be highlighted in specific sections throughout this text.

Deoxyribonucleic Acid and the Genome All organisms possess a scheme to transmit the essential information containing the genetic make-up of the species through successive generations. In human cells, 23 pairs of chromosomes are present, each pair of which contains a 2

unique sequence and therefore unique genetic information. In the human genome, about 6 × 109 nucleotides, or 3 × 109 pairs of nucleotides, associate in the double helix. All the specificity of DNA is determined by the base sequence, and this sequence is stored in complementary form in the double-helical structure, which facilitates correction of sequence errors and provides a mechanistic basis for replication of the information during cell division. Each DNA strand serves as a template for replication, which is accomplished by the action of DNA-dependent polymerases that unwind the double-helical DNA and copy each single strand with remarkable fidelity. All cell types except for gametocytes contain this duplicate, diploid number of genetic units, one half of which is referred to as a haploid number. The genetic information contained in chromosomes is separated into discrete functional elements known as genes. A gene is defined as a unit of base sequence that usually, but with rare exceptions, encodes a specific polypeptide sequence. New evidence suggests that small, noncoding RNAs play critical roles in expression of this essential information. An estimated 30,000 genes are present in the human haploid genome, and these are interspersed among regions of sequence that do not code for protein and whose function is as yet unknown. For example, noncoding RNAs (e.g., transfer RNA [tRNA], ribosomal RNA [rRNA], and other small RNAs) act as components of enzyme complexes such as the ribosome and spliceosome. The average chromosome contains 3000 to 5000 genes, and these range in size from about 1 kilobase (kb) to 2 megabases (Mb).

Ribonucleic Acid Synthesis Transcription, or RNA synthesis, is the process for transferring information contained in nuclear DNA to an intermediate molecular species known as messenger RNA (mRNA).

Chapter 1—Molecular Basis of Human Disease

Enhancer Silencer

RNA polymerase

Exon 5'

Intron

Exon

Intron 3'

Promoter region Transcription start site

Figure 1-1 Transcription. Genomic DNA is shown with enhancer and silencer sites located 5′ upstream of the promoter region, to which RNA polymerase is bound. The transcription start site is shown downstream of the promoter region, and this site is followed by exonic sequences interrupted by intronic sequences. The former sequences are transcribed ad seriatim (i.e., one after another) by the RNA polymerase.

Two biochemical differences distinguish RNA from DNA: (1) the polymeric backbone is made up of ribose rather than deoxyribose sugars linked by phosphodiester bonds, and (2) the base composition is different in that uracil is substituted for thymine. RNA synthesis from a DNA template is performed by three types of DNA-dependent RNA polymerases, each a multi-subunit complex with distinct nuclear location and substrate specificity. RNA polymerase I, located in the nucleolus, directs the transcription of genes encoding the 18S, 5.8S, and 28S ribosomal RNAs, forming a molecular scaffold with both catalytic and structural functions within the ribosome. RNA polymerase II, located in the nucleoplasm instead of the nucleoli, primarily transcribes precursor mRNA transcripts and small RNA molecules. The carboxyl-terminus of RNA polymerase II is uniquely modified with a 220-kD protein domain, the site of enzymatic regulation by protein phosphorylation of critical serine and threonine residues. All tRNA precursors and other rRNA molecules are synthesized by RNA polymerase III in the nucleoplasm. RNA polymerases are synthesized from precursor trans cripts that must first be cleaved into subunits before further processing and assembling with ribosomal proteins into macromolecular complexes. Ribosomal architectural and structural integrity are derived from the secondary and tertiary structures of rRNA, which assume a series of folding patterns containing short duplex regions. Precursors of tRNA in the nucleus undergo the removal of the 5′ leader region, splicing of an internal intron sequences, and modification of terminal residues. Precursors of mRNA are produced in the nucleus by the action of DNA-dependent RNA polymerase II, which copies the antisense strand of the DNA double helix to synthesize a single strand of mRNA that is identical to the sense strand of the DNA double helix in a process called transcription (Fig. 1-1). The initial, immature mRNA first undergoes modification at both the 5′ and 3′ ends. A special nucleotide structure called the cap is added to the 5′ end, which functions to increase binding to the ribosome and enhance translational efficiency. The 3′ end undergoes modification by nuclease cleavage of about 20 nucleotides, followed by the addition of a length of polynucleotide sequence containing a uniform stretch of adenine bases, the so-called poly A tail that stabilizes the mRNA. In addition to these changes that uniformly occur in all mRNAs, other, more selective modifications can also occur.

MET CYS

PRO

THR

Anticodon GGG

UGC AUG

UCG

ACG Codon

CCC

UCG

AUU

GUA

Open reading frame

Figure 1-2 Translation. The open reading frame of a mature messenger RNA is shown with its series of codons. Transfer RNA molecules are shown with their corresponding anticodons, charged with their specific amino acid. A short, growing polypeptide chain is depicted. A, adenine; C, cytosine; CYS, cysteine; G, guanine; MET, methionine; PRO, proline; THR, threonine; U, uracil.

Because each gene contains both exonic and intronic sequences and the precursor mRNA is transcribed without regard for exon-intron boundaries, this immature message must be edited in such a way that splices all exons together in appropriate sequence. The process of splicing, or removing intronic sequences to produce the mature mRNA, is an exquisitely choreographed event that involves the intermediate formation of a spliceosome, a large complex consisting of small nuclear RNAs and specific proteins, which contains a loop or lariat-like structure that includes the intron targeted for removal. Only after splicing, a catalytic process requiring adenosine triphosphate hydrolysis, has concluded is the mature mRNA able to transit from the nucleus into the cytoplasm, where the encoded information is translated into protein. Alternative splicing is a process for efficiently generating multiple gene products often dictated by tissue specificity, developmental expression, and pathologic state. Gene splicing allows the expression of multiple isoforms by expanding the repertoire for molecular diversity. An estimated 30% of genetic diseases in humans arise from defects in splicing. The resulting mature mRNA then exits the nucleus to begin the process of translation or conversion of the base code to polypeptide (Fig. 1-2). Alternative splicing pathways (i.e., alternative exonic assembly pathways) for specific genes also serve at the level of transcriptional regulation. The discovery of catalytic RNA, the capacity for self-directed internal

Section I—Introduction to Molecular Medicine 3' OH A C C Phosphorylated 5' terminus

Amino acid– attachment site

5' p

TψC loop

DHU loop U

UH2

C T Ψ

G "Extra arm" (variable)

Anticodon loop

Figure 1-3 Secondary structure of transfer RNA (tRNA). The structure of each tRNA serves as an adapter molecule that recognizes a specific codon for the amino acid to be added to the polypeptide chain. About one half the hydrogen-bonded bases of the single chain of ribonucleotides are shown paired in double helices like a cloverleaf. The 5′ terminus is phosphorylated, and the 3′ terminus contains the hydroxyl group on an attached amino acid. The anticodon loop is typically located in the middle of the tRNA molecule. C, cytocide; DHU, dihydroxyuridine; G, guanine; UH2, dihydrouridine; ψ, pseudouridine; T, ribothymidine; U, uracil. (Data from Berg JM, Tymoczko JL, Strayer JL: Berg, Tymoczko and Stryer’s Biochemistry, 5th ed. New York, WH Freeman, 2006.)

excision and repair, has advanced the current view that RNA per se serves both as a template for translation of the genetic code and, simultaneously, as an enzyme (see “Transcriptional Regulation” later in this chapter). Protein synthesis, or translation of the mRNA code, occurs on ribosomes, which are macromolecular complexes of proteins and rRNA located in the cytoplasm. Translation involves the conversion of the linear code of a triplet of bases (i.e., the codon) into the corresponding amino acid. A fourbase code generates 64 possible triplet combinations (4 × 4 × 4), and these correspond to 20 different amino acids, many of which are encoded by more than one base triplet. To decode mRNA, an adapter molecule (tRNA) recognizes the codon in mRNA through complementary base pairing with a three-base anticodon that it bears; in addition, each tRNA is charged with a unique amino acid that corresponds to the anticodon (Fig. 1-3). Translation on the mRNA template proceeds without punctuation of the non-overlapping code with the aid of rRNA on an assembly machine, termed ribosomes—essen-

tially a polypeptide polymerase. At least one tRNA molecule exists for each 20 amino acids, although degeneracy in the code expands the number of available tRNA molecules, mitigates the chances of premature chain termination, and ameliorates the potential deleterious consequences of single-base mutations. The enzymatic activity of the ribosome then links amino acids through the synthesis of a peptide bond, releasing the tRNA in the process. Consecutive linkage of amino acids in the growing polypeptide chain represents the terminal event in the conversion of information contained within the nuclear DNA sequence into mature protein (DNA → RNA → protein). Proteins are directly responsible for the form and function of an organism. Thus, abnormalities in protein structure or function brought about by changes in primary amino acid sequence are the immediate precedent cause of changes in phenotype, adverse forms of which define a disease state. Inhibition of RNA synthesis is a well-recognized mechanism of specific toxins and antibiotics. Toxicity from the ingestion of the poisonous mushroom (Amanita phalloides), for example, leads to the release of the toxin α-amanitin, a cyclic octapeptide that inhibits the RNA Pol II and blocks elongation of RNA synthesis. The antibiotic actinomycin D binds with high affinity to double-helical DNA and inter colates between base pairs, precluding access of DNAdependent RNA polymerases and the selective inhibition of transcription. Several major antibiotics function through inhibition of translation. For example, the aminoglycoside antibiotics function through the disruption of the mRNAtRNA codon-anticodon interaction, whereas erythromycin and chloramphenicol inhibit peptide bond formation.

Control of Gene Expression OVERVIEW The timing, duration, localization, and magnitude of gene expression are all important elements in the complex tapestry of cell form and function governed by the genome. Gene expression represents the flow of information from the DNA template into mRNA transcripts and the process of translation into mature protein. Four levels of organization involving transcription factors, RNAs, chromatin structure, and epigenetic factors are increasingly recognized to orchestrate gene expression in the mammalian genome. Transcriptional regulators bind to specific DNA motifs that positively or negatively control the expression of neighboring genes. The information contained in the genome must be transformed into functional units of either RNA or protein products. How DNA is packed and modified represents additional modes of gene regulation by disrupting the access of transcription factors from DNAbinding motifs. In the postgenomic era, the challenge is to understand the architecture by which the genome is organized, controlled, and modulated. Transcription factors, chromatin architecture, and modifications of nucleosomal organization make up the major mechanisms of gene regulation in the genome.

Chapter 1—Molecular Basis of Human Disease

TRANSCRIPTIONAL REGULATION The principal regulatory step in gene expression occurs at the level of gene transcription. A specific DNA-dependent RNA polymerase performs the transcription of information contained in genomic DNA into mRNA transcripts. Trans cription begins at a proximal (i.e., toward the 5′ end of the gene) transcription start site, containing nucleotide sequences that influence the rate and extent of the process (see Fig. 1-1). This region is known as the promoter region of the gene and often includes an element of sequence rich in adenine and thymine (the TATA box) along with other sequence motifs within about 100 bases of the start site. These regions of DNA that regulate transcription are known as cis-acting regulatory elements. Some of these regulatory regions of promoter sequence bind proteins known as transacting factors, or transcription factors, which are themselves encoded by other genes. The cis-acting regulatory sequences to which transcription factors bind are often referred to as response elements. Families of transcription factors have been identified and are often described by unique aspects of their predicted protein secondary structure, including helix-turnhelix motifs, zinc-finger motifs, and leucine-zipper motifs. Transcription factors make up an estimated 3% to 5% of the protein-coding products of the genome. In addition to gene-promoter regions, enhancer sites are distinct from promoter sites in that they can exist at distances quite remote from the start site, either upstream or downstream (i.e., beyond the 3′ end of the gene), and without clear orientation requirements. Trans-acting factors bind to these enhancer sites and are believed to alter the tertiary structure or conformation of the DNA in a manner that facilitates the binding and assembly of the transcription-initiation complex at the promoter region, perhaps in some cases by forming a broad loop of DNA in the process. Biochemical modification of select promoter or enhancer sequences, such as methylation of CpG-rich sequences (cytosine-phosphate-guanine), can also modulate transcription; methylation typically suppresses transcription. The terms silencer and suppressor elements refer to cis-acting nucleotide sequences that reduce or shut off gene transcription and do so through association with trans-acting factors that recognize these specific sequences. Regulation of transcription is a complex process that occurs at several levels; importantly, the expression of many genes is regulated to maintain high basal levels, which are known as housekeeping or constitutively expressed genes. They typically yield protein products that are essential for normal cell function or survival and thus must be maintained at a specific steady-state concentration under all circumstances. Many other genes, in contrast, are not expressed or are only modestly expressed under basal conditions; however, with the imposition of some stress or exposure of the cell to an agonist that elicits a cellular response distinct from that of the basal state, the expression of these genes is induced or enhanced. For example, the heat shock protein genes encoding stress proteins are rapidly induced in response to diverse pathophysiologic stimuli (e.g., oxidative stress, heavy metals, inflammation) in most cells and organisms. The increased heat shock protein expression is complementary to the basal level of heat shock proteins whose functions as molecular chaperones play key roles during protein synthesis to prevent protein misfolding, increase protein translocation, and

accelerate protein degradation. These adaptive responses often mediate changes in phenotype that are homeostatically protective to the cell or organism.

MICRORNAS AND GENE REGULATION Less is currently known about the determinants of translational regulation than is known about transcriptional regulation. The recent discovery and identification of small RNAs (21 to 24 mer), termed microRNAs (miRNAs), adds further complexity to the regulation of gene expression within the eukaryotic genome. First discovered in worms more than 10 years ago, miRNAs are conserved noncoding strands of RNA that bind to the 3′-untranslated regions of target mRNAs, enabling gene silencing of protein expression at the translational level. Gene-encoding miRNAs exhibit tissue-specific expression and are interspersed in regions of the genome unrelated to known genes. Transcription of miRNAs proceeds in multiple steps from sites under the control of an mRNA promoter. RNA polymerase II transcribes the precursor miRNA, termed primary miRNA (primiRNA), containing 5′ caps and 3′ poly (A) tails. In the nucleus, the larger primiRNAs of 70 nucleotides form an internal hairpin loop, embedding the miRNA portion that undergoes recognition and subsequent excision by double-stranded RNA-specific ribonuclease, termed Drosha. Gene expression is silenced by the effect of miRNA on nascent RNA molecules targeted for degradation. Because translation occurs at a fairly invariant rate among all mRNA species, the stability or half-life of a specific mRNA also serves as another point of regulation of gene expression. The 3′-untranslated region of mRNAs contains regions of sequence that dictate the susceptibility of the message to nuclease cleavage and degradation. Stability appears to be sequence specific and, in some cases, dependent on transacting factors that bind to the mRNA. The mature mRNA contains elements of untranslated sequence at both the 5′ and 3′ ends that can regulate translation. Beginning in the organism’s early development, miRNAs may facilitate much more intricate ways for the regulation of gene expression, as have been shown for germline production, cell differentiation, proliferation, and organogenesis. Because recent studies have implicated the expression of miRNAs in brain development, cardiac organogenesis, colonic adenocarcinoma, and viral replication, this novel mechanism for gene silencing has potential therapeutic roles for congenital heart defects, viral disease, neurodegeneration, and cancer.

CHROMATIN REMODELING AND GENE REGULATION Both the size and complexity of the human genome with 23 chromosomes, ranging in size between 50 and 250Mb, pose formidable challenges for transcription factors to exert the specificity of DNA-binding properties in gene regulation. Control of gene expression also takes place in diverse types of cells, often with exquisite temporal and spatial specificity throughout the life span of the organism. In eukaryotic cells, the genome is highly organized into densely packed nucleic acid DNA- and RNA-protein structures, termed chromatin.

Section I—Introduction to Molecular Medicine

Figure 1-4 Schematic representation of a nucleosome. Rectangular blocks represent the DNA strand wrapped around the core that consists of eight histone proteins. Each histone has a protruding tail that can be modified to repress or activate transcription. (Adapted from Berg JM, Tymoczko JL, Strayer JL: Berg, Tymoczko and Stryer’s Biochemistry, 5th ed. New York, WH Freeman, 2006.)

The building blocks of chromatin are called histones, a family of small basic proteins that occupy one half of the mass of the chromosome. Histones derive their basic properties from the high content of basic amino acids, arginine, and lysine. Five major types of histones—H1, H2A, H2B, H3, and H4—have evolved to form complexes with the DNA of the genome. Two pairs each of the four types of histones form a protein core, the histone octomer, which is wrapped by 200 base pairs of DNA to form the nucleosome (Fig. 1-4). The core proteins within the nucleosomes have protruding amino-terminal ends, exposing critical lysine and arginine residues for covalent modification. Further DNA condensation is achieved as higher-order structure is imparted on the chromosomes. The nucleosomes are further compacted in layered stacks with a left-handed superhelix resulting in negative supercoils that provide the energy for DNA strand separation during replication. Condensation of DNA in chromatin precludes the access of regulatory molecules such as transcription factors. Reversal of chromatin condensation, on the other hand, typically occurs in response to environmental and other developmental signals in a tissue-dependent manner. Promoter sites undergoing active transcription, as well as relaxation of chromatin structure, that become susceptible to enzymatic cleavage by nonspecific DNAase I are called hypersensitive sites. Transcription factors on promoter sites may gain access by protein-protein interactions to enhancer elements containing tissue-specific proteins at remote sites, several thousand bases away, resulting in transcription activation or repression.

EPIGENETIC CONTROL OF GENE EXPRESSION Complex regulatory networks revolve around transcription factors, nucleosomes, chromatin structure, and epigenetic markings. Epigenetics refers to heritable changes in gene

expression without changes in the DNA sequence. Such examples include DNA methylation, gene silencing, chromatin remodeling, and X-chromosome inactivation. This form of inheritance involves the alterations in gene function without changes in DNA sequence. Chemical marking of DNA methylation is both cell specific and developmentally regulated. Methylation of the 5′ CpG dinucleotide by specific methyl transferases, which occurs in 70% of the mammalian genome, is another mechanism of gene regulation. Steric hindrance from the bulky methyl group of 5′ methylcytosine precludes occupancy by transcription factors that stimulate or attenuate gene expression. Most genes are found in CpG islands, reflecting sites of gene activity across the genome. In an analogous manner, modifications of histone by phosphorylation, methylation, ubiquination, and acetylation are transmitted and reestablished in an inheritable manner. It is conceivable that other epigenetic mechanisms do not involve genomic modifications of DNA. For example, modification of the gene encoding the estrogen receptor α has been implicated in gene silencing at 5mC sites of multiple downstream targets in breast cancer cells. Powerful new approaches are being developed to examine feedback and feed-forward loops in transmission of epigenetic markings. The concept that dynamic modifications (e.g., DNA methylation and acetylation) of histones or epigenesis contribute, in part, to tumorigenic potential for progression has already been translated into current therapies. Histone acetyltransferases (HATs) and histone deacetyltransferases (HDACs) play antagonistic roles in the addition and removal of acetylation in the genome. Furthermore, genome-wide analysis of HATs and HDACs is beginning to provide important insights into complex modes of gene regulation. Several inhibitors of histone deacetylases, with a range of biochemical and biologic activities, are being developed and tested as anticancer agents in clinical trial. Phase I clinical trials have suggested these drugs are well tolerated. In general, the inhibition of deacetylase remodels chromatin assembly and reactivates transcription of the genome. Because the mechanisms of actions of HDACs extend to apoptosis, cell cycle control, and cellular differentiation, current clinical trials are seeking to determine the efficacy of these novel reagents in the drug compendium for human cancers.

Genetic Sequence Variation, Population Diversity, and Genetic Polymorphisms A stable, heritable change in DNA is defined as a mutation. This strict contemporary definition does not depend on the functional relevance of the sequence alteration and implicates a change in primary DNA sequence. Considered in historical context, mutations were first defined on the basis of identifiable changes in the heritable phenotype of an organism. As biochemical phenotyping became more precise in the mid-20th century, investigators demonstrated that many proteins exist in more than one form in a population, and these forms were viewed as a consequence of variations

Chapter 1—Molecular Basis of Human Disease in the gene coding for that protein (i.e., allelic variation). With advances in DNA-sequencing methods, the concept of mutation evolved from one that could be appreciated only by identifying differences in phenotype to one that could precisely be defined at the level of changes in the structure of DNA. Although most mutations are stably transmitted from parents to offspring, some are genetically lethal and thus cannot be passed on. In addition, the discovery of regions of the genome that contain sequences that repeat in tandem a highly variable number of times (tandem repeats) suggests that some mutations are less stable than others. These tandem repeats are further described later in this section. The molecular nature of mutations is varied (Table 1-1). A mutation can involve the deletion, insertion, or substitution of a single base, all of which are referred to as point mutations. Substitutions can be further classified as silent when the amino acid encoded by the mutated triplet does not change, as missense when the amino acid encoded by the mutated triplet changes, and as nonsense when the mutation leads to premature termination of translation (stop codon). On occasion, point mutations can alter the processing of precursor mRNA by producing alternate splice sites or eliminating a splice site. When a single- or double-base deletion or insertion occurs in an exon, a frameshift mutation results, usually leading to premature termination of translation at a now in-frame stop codon. The other end of the spectrum of mutations includes large deletions of an entire gene or a set of contiguous genes; deletion, duplication, and translocation of a segment of one chromosome to another; or duplication or deletion of an entire chromosome. Such chromosomal mutations play a large role in the development of many cancers. Each individual possesses two alleles for any given gene locus, one from each parent. Identical alleles define homozy-

Table 1-1 Molecular Basis of Mutations Type

Examples

Point Mutations Deletion Substitution Silent Missense Nonsense

α-Thalassemia, polycystic kidney disease Cystic fibrosis Sickle cell anemia, polycystic kidney disease, congenital long QT syndrome Cystic fibrosis, polycystic kidney disease

Large Mutations (Gene or Gene Cluster) Deletion Insertion Duplication Inversion Expanding triplet

Duchenne muscular dystrophy Factor VIII deficiency (hemophilia A) Duchenne muscular dystrophy Factor VIII deficiency Huntington disease

Very Large Mutation (Chromosomal Segment or Chromosome) Deletion Duplication Translocation

Turner syndrome (45,X) Trisomy 21 XX male [46,X; t(X;Y)]*

*Translocation onto an X chromosome of a segment of a Y chromosome that bears the locus for testicular differentiation.

gosity and nonidentical alleles define heterozygosity for any gene locus. The heritability of these alleles follows typical mendelian rules. With a clearer understanding of the molecular basis of mutations and of allelic variation, their distribution in populations can now be analyzed precisely by following specific DNA sequences. Differences in DNA sequences studied within the context of a population are referred to as genetic polymorphisms, and these polymorphisms underlie the diversity observed within a given species and among species. Despite the high prevalence of benign polymorphisms in a population, the occurrence of harmful mutations is comparatively rare because of selective pressures that eliminate the most harmful mutations from the population (lethality) and the variability within the genomic sequence to polymorphic change. Some portions of the genome are remarkably stable and free of polymorphic variation, whereas other portions are highly polymorphic, the persistence of variation within which is a consequence of the functional benignity of the sequence change. In other words, polymorphic differences in DNA sequence between individuals can be divided into those producing no effect on phenotype, those causing benign differences in phenotype (i.e., normal genetic variation), and those producing adverse consequences in phenotype (i.e., mutations). The last group can be further subdivided into the polymorphic mutations that alone are able to produce a functionally abnormal phenotype such as monogenic disease (e.g., sickle cell anemia) and those that alone are unable to do so but in conjunction with other mutations can produce a functionally abnormal phenotype (complex disease traits [e.g., essential hypertension]). Polymorphisms are more common in noncoding regions of the genome than they are in coding regions, and one common type of these involves the tandem repetition of short DNA sequences a variable number of times. If these tandem repeats are long, they are termed variable number tandem repeats; if these repeats are short, they are termed short tandem repeats (STRs). During mitosis, the number of tandem repeats can change, and the frequency of this kind of replication error is high enough to make alternative lengths of the tandem repeats common in a population. However, the rate of change in length of the tandem repeats is low enough to make the size of the polymorphism useful as a stable genotypic trait in families. In view of these features, polymorphic tandem repeats are useful in determining the familial heritability of specific genomic loci. Polymorphic tandem repeats are sufficiently prevalent along the entire genomic sequence, enabling them to serve as genetic markers for specific genes of interest through an analysis of their linkage to those genes during crossover and recombination events. Analyses of multiple genetic polymorphisms in the human genome reveal that a remarkable variation exists among individuals at the level of the sequence of genomic DNA (genotyping). Single-nucleotide polymorphism (SNP), the most common variant, differs by a single base between chromosomes on any given stretch of DNA sequence (Fig. 1-5). From genotyping of the world’s representative population, 10 million variants (one site per 300 bases) are estimated to make up 90% of the common SNP variants in the population, with the rare variants making up the remaining 10%. With each generation of a species, the frequency of polymorphic changes in a gene is 10−4 to 10−7.

Section I—Introduction to Molecular Medicine

Figure 1-5 Single nucleotide polymorphisms (SNPs), haplotypes, and tag SNPs. A stretch of mostly identical DNA on the same chromosome is shown from four different individuals. SNP refers to the variation of the three bases shown in DNA region. The combination of nearby SNPs defines a haplotype. Tag SNPs are useful tools shown (C) for genotyping four unique haplotypes from the 20 haplotypes (B). (Adapted from International HapMap Consortium: The International HapMap Project. Nature 426: 789-796, 2003.)

Thus, in view of the number of genes in the human genome, between 0.5% and 1.0% of the base sequence of the human genome is polymorphic. In this context, the new variant can be traced historically to the surrounding alleles on the chromosomal background present at the time of the mutational event. A haplotype is a specific set or combination of alleles on a chromosome or part of a chromosome (see Fig. 1-5). When parental chromosomes undergo crossover, new mosaic haplotypes, containing additional mutations, are created from such recombinations. SNP alleles within haplotypes can be co-inherited in association with other alleles in the population, termed linkage disequilibrium (LD). The association between two SNPs will decline with increasing distance, enabling patterns of LD to be decided from the proximity of nearby SNPs. Conversely, a few well-selected SNPs are often sufficient to predict the location of other common variants in the region. Haplotypes associated with a mutation are expected to become common by recombination in the general population over thousands of generations. In contrast, genetic mapping with LD departs from traditional mendelian genetics by using the entire human population as a large family tree without an established pedigree. Of the possible 10 million variants, the International HapMap Project and the Perlegen private venture have deposited more than 8 million variants comprising the public human SNP map from more than 341 people representing different population samples. The SNPs distributed across the genome of unrelated individuals provide a sufficiently robust sample set for statistical associations to be drawn between genotypes and modest phenotypes. A mutation can now be defined as a specific type of allelic polymorphism that causes a functional defect in a cell or organism.

The causal relationship between monogenic diseases with well-defined phenotypes that co-segregate with the disease requires only a small number of affected individuals compared with unaffected control individuals. In contrast, complex disorders (e.g., diabetes, hypertension, cancer) will necessitate the combinatorial effects of environmental factors and genes with subtle effects. Only through searching for variations in genetic frequency between patients and the general population can the causation of disease be discerned. In the postgenomic era, gene mapping entails the statistical association with the use of LD and high-density genetic maps that span thousands to 100,000 base pairs. To enable comprehensive association studies to become routine in clinical practice, inexpensive genotyping assays and denser maps with all common polymorphisms must be linked to all possible manifestations of the disease. Longitudinal studies of the HapMap and Perlegen cohorts will determine the effects of diet, exercise, environmental factors, and family history on future clinical events. Without similar approaches on securing adequate sample sizes and datasets, the promise of genetic population theory will not overcome the inherent limitations of linking human sequence variation with complex disease traits.

Gene Mapping and the Human Genome Project The process of gene mapping involves identifying the relative order and distance of specific loci along the genome. Maps can be of two types: genetic and physical. Genetic maps identify the genomic location of specific genetic loci by a statistical analysis based on the frequency of recombina-

Chapter 1—Molecular Basis of Human Disease

C Figure 1-6 Crossing over and recombination. A, Two haploid chromosomes are shown, one from each parent (red and blue) with two genomic loci denoted by the circles and squares. B, Crossing over of one haploid chromosome from each parent. C, Resulting recombination of chromosomal segments now redistributes one haploid locus (squares) from one diploid pair to another.

tion events of the locus of interest with other known loci. Physical maps identify the genomic location of specific genetic loci by a direct measurement of the distance along the genome at which the locus of interest is located in relation to one or more defined markers. The precise location of genes on a chromosome is important for defining the likelihood that a portion of one chromosome will interchange, or cross over, with the corresponding portion of its complementary chromosome when genetic recombination occurs during meiosis (Fig. 1-6). During meiotic recombination, genetic loci or alleles that have been acquired from one parent interchange with those acquired from the other parent to produce new combinations of alleles, and the likelihood that alleles will recombine during meiosis varies as a function of their linear distance from one another in the chromosomal sequence. This recombination probability or distance is commonly quantitated in centimorgans (cM): 1cM is defined as the chromosomal distance over which there is a 1% chance that two alleles will undergo a crossover event during meiosis. Crossover events serve as the basis for mixing parental base sequences during development and, thereby, promoting genetic diversity among offspring. Analysis of the tendency

for specific alleles to be inherited together indicates that the recombination distance in the human genome is about 3000cM. Identifying the gene or genes responsible for a specific polygenic disease phenotype requires an understanding of the topographic anatomy of the human genome, which is inextricably linked to interactions with the environment. The Human Genome Project, first proposed in 1985, represented an international effort to determine the complete nucleotide sequence of the human genome, including the construction of its detailed genetic, physical, and transcript maps, with identification and characterization of all genes. This foray into large-scale biology was championed by Nobel Laureate James Watson as the defining moment in his lifetime for witnessing the path from the double helix to the sequencing of 3 billion bases of the human genome, paving the way for understanding human evolution and harnessing the benefits for human health. Among the earliest achievements of the Human Genome Project were the development of 1-cM resolution maps, each containing 3000 markers, and the identification of 52,000 sequenced tagged sites. For functional analysis on a genome-wide scale, major technologic advances were made, including as high-throughput oligonucleotide synthesis, normalized and subtracted complementary DNA (cDNA) libraries, and DNA micro-arrays. In 1998, the Celera private venture proposed a similar goal as the Human Genome Project using a revolutionary approach, termed shotgun sequencing, to determine the sequence of the human genome (Web Movie 1-1). The shotgun sequencing method was designed for random large-scale sequencing and subsequent alignment of sequenced segments using computational and mathematic modeling. In the end, the Human Genome Project, in collaboration with the Celera private venture, produced a refined map of the entire human genome in 2001. Because of the differences in genomic sequence that arise as a consequence of normal biologic variations or sequence polymorphisms, the resulting restriction fragment length polymorphisms (RFLPs) differ among individuals and are inherited according to mendelian principles. These polymorphisms can serve as genetic markers for specific loci in the genome. One of the most useful types of RFLP for localization of genetic loci within the genome is that produced by tandem repeats of sequence. Tandem repeats arise through slippage or stuttering of the DNA polymerase during replication in the case of STRs; longer variations arise through unequal crossover events. STRs are distributed throughout the genome and are highly polymorphic. Of importance is that these markers have two different alleles at each locus that are derived from each parent; thus, the origins of the two chromosomes can be discerned through this analysis. The use of highly polymorphic tandem repeats that occur throughout the genome as genomic markers has provided a basis for mapping specific gene loci through establishing the association or linkage with select markers. Linkage analysis is predicated on a simple principle: the likelihood that a crossover event will occur during meiosis decreases the closer the locus of interest is to a given marker. The extent of genetic linkage can be ascertained for any group of loci, one of which may contain a disease-producing mutation (Fig. 1-7).

Section I—Introduction to Molecular Medicine

A M

B WT

A M

B WT

Figure 1-7 Linkage analysis. Analysis of the association (genomic contiguity) of a mutation (M) and a polymorphic allelic marker (A) shows close linkage in that the mutation segregates with the A allele, whereas the wild-type gene locus (WT) associates with the B allele.

Identifying Mutant Genes Deducing the identity of a specific gene sequence believed to cause a specific human disease requires that mutations in the gene of interest be identified. If the gene believed to be responsible for the disease phenotype is known, its sequence can be determined by conventional cloning and sequencing strategies, and the mutation can be identified. A variety of techniques are currently available for detecting mutations. Mutations that involve insertion or deletion of large segments of DNA can be detected by Southern blot, in which the isolated DNA is annealed to a radioactively labeled fragment of cDNA sequence. Prior incubation of the DNA with a specific restriction endonuclease cleaves the DNA sequence of interest at specific sites to produce smaller fragments that can be monitored by agarose gel electrophoresis. Shifts in mobility on the gel in comparison with wild-type sequence become apparent as a function of changes in the molecular size of the fragment. Alternatively, the polymerase chain reaction (PCR) can be used to identify mutations (Web Movie 1-2). In this approach, small oligonucleotides (20 to 40 bases in length), which are complementary to regions of DNA that bracket the sequence of interest and are complementary to each strand of the double-stranded DNA, are synthesized and serve as primers for the amplification of the DNA sequence of interest. These primers are added to the DNA solution. The temperature of the solution is increased to dissociate the individual DNA strands and is then reduced to permit annealing of the primers to their complementary template target sites. A thermostable DNA polymerase is included in the reaction to synthesize new DNA in the 5′-to3′ direction from the primer annealing sites. The temperature is then increased to dissociate duplex structures, after which it is reduced, enabling another cycle of DNA synthesis to occur. Several temperature cycles (usually up to 40) are used to amplify progressively the concentration of the

sequence of interest, which can be identified as a PCR product by agarose gel electrophoresis with a fluorescent dye. The product can be isolated and sequenced to identify the suggested mutation. If the gene is large and the site of the mutation is unknown (especially if it is a point mutation), other methods can be used to identify the likely mutated site in the exonic sequence. One commonly used approach involves scanning the gene sequence for mutations that alter the structural conformation of short complexes between parent DNA and PCR products, leading to a shift in mobility on a nondenaturing agarose gel (i.e., single-strand conformational polymorphism). A single-base substitution or deletion can change the conformation of the complex in comparison with wildtype complexes and yield a shift in mobility. Sequencing this comparatively small region of the gene then facilitates precise identification of the mutation. When the gene believed to cause the disease phenotype is unknown, when its likely position on the genome has not been identified, or when only limited mapping information is available, a candidate gene approach can be used to identify the mutated gene. In this strategy, potential candidate genes are identified on the basis of analogy to animal models or by analysis of known genes that map to the region of the genome for which limited information is available. The candidate gene is then analyzed for potential mutations. Regardless of the approach used, mutations identified in candidate genes should always be correlated with functional changes in the gene product because some mutations could be functionally silent, representing a polymorphism without phenotypic consequences. Functional changes in the gene product can be evaluated through the use of cell-culture systems to assess protein function by expressing the mutant protein through transiently transfecting the cells with a vector that carries the cDNA coding for the gene of interest and incorporating the mutation of interest. Alternatively, unique animal models can be developed in which the mutant gene is incorporated in the male pronucleus of oocytes taken from a super-ovulating impregnated female. This union produces an animal that overexpresses the mutant gene; that is, it produces a transgenic animal, an animal with more than the usual number of copies of a given gene, or an animal in which the gene of interest is disrupted and the gene product is not synthesized (i.e., a gene knockout animal or an animal with one half [heterozygote] or none [homozygote] of the usual number of a given gene).

MOLECULAR DIAGNOSTICS The power of molecular techniques extends beyond their use in defining the precise molecular basis of an inherited disease. By exploiting the exquisite sensitivity of PCR to amplify rare nucleic acid sequences, it is possible to diagnose rapidly a range of infectious diseases for which unique sequences are available. In particular, infections caused by fastidious or slow-growing organisms can now be rapidly diagnosed, similar to the case for Mycobacterium tuberculosis. The presence of genes conferring resistance to specific antibiotics in microorganisms can also be verified by PCR techniques. The sequencing of the entire genome of organisms such as Escherichia coli, M. tuberculosis, and Treponema pallidum now offers unparalleled opportunities to monitor

Chapter 1—Molecular Basis of Human Disease the epidemiologic structures of infections, follow the course of acquired mutations, tailor antibiotic therapies, and develop unique gene-based therapies (see later) for infectious agents for which conventional antibiotic therapies are ineffective or marginally effective. The application of molecular methods to human genetics has clearly revolutionized the field. Through the use of approaches that incorporate linkage analysis and PCR, simple point mutations can be precisely localized and characterized. At the other end of the spectrum of genetic changes that underlie disease, chromosomal translocations, deletions, or duplications can be identified by conventional cytogenetic methods. Large deletions that can incorporate many kilobase pairs and many genes can now be visualized with fluorescent in situ hybridization (FISH), a technique in which a segment of cloned DNA is labeled with a fluorescent tag and hybridized to chromosomal DNA. With the deletion of the segment of interest from the genome, the chromosomal DNA fails to fluoresce in the corresponding chromosomal location. Advances in molecular medicine have also revolutionized the approach to the diagnosis and treatment of neoplastic diseases, as well as the understanding of the mechanisms of carcinogenesis. According to current views, a neoplasm arises from the clonal proliferation of a single cell that is transformed from a regulated, quiescent state into an unregulated growth phase. DNA damage accumulates in the parental tumor cell as a result of either exogenous factors (e.g., radiation exposure) or heritable determinants. In early phases of carcinogenesis, certain genomic changes may impart intrinsic genetic instability that increases the likelihood of additional damage. One class of genes that becomes activated during carcinogenesis is oncogenes, which are primordial genes that normally exist in the mammalian genome in an inactive (proto-oncogene) state but, when activated, promote unregulated cell proliferation through activation of specific intracellular signaling pathways. Molecular methods based on the acquisition of specific tumor markers and unique DNA sequences that result from oncogenetic markers of larger chromosomal abnormalities (i.e., translocations or deletions that promote oncogenesis) are now broadly applied to the diagnosis of malignancies. These methods can be used to establish the presence of specific tumor markers and oncogenes in biopsy specimens, to monitor the presence or persistence of circulating malignant cells after completion of a course of chemotherapy, and to identify the development of genetic resistance to specific chemotherapeutic agents. In addition, through the use of conventional linkage analysis and candidate gene approaches, future studies will enable the identification of individuals with a heritable predisposition to malignant transformation. Many of these specific topics are discussed in later chapters. The advent of gene chip technologies or expression arrays has revolutionized molecular diagnostics and has begun to clarify the pathobiologic structures of complex diseases. These methods involve labeling the cDNA generated from the entire pool of mRNA isolated from a cell or tissue specimen with a radioactive or fluorescent marker and annealing this heterogeneous population of polynucleotides to a solidphase substrate to which many different polynucleotides of known sequence are attached. The signals from the

labeled cDNA strands bound to specific locations on the array are monitored, and the relative abundance of particular sequences is compared with that from a reference specimen. Using this approach, micro-array patterns can be used as molecular fingerprints to diagnose a particular disease (i.e., type of malignancy and its susceptibility to treatment and prognosis) as well as to identify the genes whose expression increases or decreases in a specific disease state (i.e., identification of disease-modifying genes). Of course, many other applications of molecular medicine techniques are available, in addition to these applications in infectious diseases and oncology. Molecular methods can be used to sort out genetic differences in metabolism that may modulate pharmacologic responses in a population of individuals (pharmacogenomics), address specific forensic issues such as paternity or criminal culpability, and approach epidemiologic analysis on a precise genetic basis.

GENES AND HUMAN DISEASE Human genetic diseases can be divided into three broad categories: (1) those that are caused by a mutation in a single gene (e.g., monogenic disorders, mendelian traits); (2) those that are caused by mutations in more than one gene (e.g., polygenic disorders, complex disease traits); and (3) those that are chromosomal in nature (Table 1-2). In all three groups of disorders, environmental factors can contribute to the phenotypic expression of the disease by modulating gene expression or unmasking a biochemical abnormality that has no functional consequences in the absence of a stimulus or stress. Classic monogenic disorders include sickle cell anemia, familial hypercholesterolemia, and cystic fibrosis. Importantly, these genetic diseases can be exclusively produced by a single specific mutation (e.g., sickle cell anemia) or by any one of several mutations (e.g., familial hypercholesterolemia, cystic fibrosis) in a given family (Pauling paradigm). Interestingly, some of these disorders evolved to protect the host. For example, sickle cell anemia evolved as protection against falciparum malaria, and cystic fibrosis developed as protection against cholera. Examples of polygenic disorders or complex disease traits include type 1 (insulin-dependent) diabetes mellitus, atherosclerotic cardiovascular disease, and essential hypertension. A common

Table 1-2 Molecular Basis of Mutations Type Monogenic Disorders Autosomal dominant Autosomal recessive X-linked One of multiple mutations

Examples Polycystic kidney disease 1, neurofibromatosis 1 β-Thalassemia, Gaucher disease Hemophilia A, Emery-Dreifuss muscular dystrophy Familial hypercholesterolemia, cystic fibrosis

Polygenic Disorders Complex disease traits

Type 1 (insulin-dependent) diabetes, essential hypertension, atherosclerotic disease, cancer

Section I—Introduction to Molecular Medicine

example of a chromosomal disorder is the presence of an extra chromosome 21 (trisomy 21). The overall frequency of monogenic disorders is about 1%. About 60% of these include polygenic disorders, which includes those with a genetic substrate that develops later in life. About 0.5% of monogenic disorders include chromosomal abnormalities. Importantly, chromosomal abnormalities are frequent causes of spontaneous abortion and malformations. Contrary to the view held by early geneticists, few phenotypes are entirely defined by a single genetic locus. Thus, monogenic disorders are comparatively uncommon; however, they are still useful as a means to understanding some basic principles of heredity. Monogenic disorders are of three types: autosomal dominant, autosomal recessive, and X-linked. Dominance and recessiveness refer to the nature of the heritability of a genetic trait and correlate with the number of alleles affected at a given locus. If a mutation in a single allele determines the phenotype, the mutation is said to be dominant; that is, the heterozygous state conveys the clinical phenotype to the individual. In contrast, if a mutation is necessary at both alleles to determine the phenotype, the mutation is said to be recessive; that is, only the homozygous state conveys the clinical phenotype to the individual. Dominant or recessive mutations can lead to either a loss or a gain of function of the gene product. If the mutation is present on the X chromosome, it is defined as X-linked (which in males can, by definition, be viewed only as dominant); otherwise, it is autosomal. The importance of identifying a potential genetic disease as inherited by one of these three mechanisms is that, if one of these patterns of inheritance is present, the disease must involve a single genomic abnormality that leads to an abnormality in a single protein. Classically identified genetic diseases are produced by mutations that affect coding (exonic) sequences. However, mutations in intronic and other untranslated regions of the genome occur that may disturb the function or expression of specific genes. Examples of diseases with these types of mutations include myotonic dystrophy and Friedreich ataxia. An individual with a dominant monogenic disorder typically has one affected parent and a 50% chance of transmitting the mutation to his or her offspring. In addition, men and women are equally likely to be affected and equally likely to transmit the trait to their offspring. The trait cannot be transmitted to offspring by two unaffected parents. In contrast, an individual with a recessive monogenic disorder typically has parents who are clinically normal. Affected parents, each heterozygous for the mutation, have a 25% chance of transmitting the clinical phenotype to their offspring but a 50% chance of transmitting the mutation to their offspring (i.e., producing an unaffected carrier). Notwithstanding the clear heritability of common monogenic disorders (e.g., sickle cell anemia), the clinical expression of the disease in an individual with a phenotype expected to produce the disease may vary. Variability in clinical expression is defined as the range of phenotypic effects observed in individuals carrying a given mutation. Penetrance refers to a smaller subset of individuals with variable clinical expression of a mutation and is defined as the proportion of individuals with a given genotype who exhibit any clinical phenotypic features of the disorder.

Three principal determinants of variability in clinical expression or incomplete penetrance of a given genetic disorder can occur: (1) environmental factors, (2) the effects of other genetic loci, and (3) random chance. Environmental factors can modulate disease phenotype by altering gene expression in several ways, including their action on trans cription factors (e.g., transcription factors that are sensitive to cell redox state [nuclear factor κB]) or cis-elements in gene promoters (e.g., folate-dependent methylation of CpG-rich regions); or by post-translationally modifying proteins (e.g., lysine oxidation). That other genes can modify the effects of disease-causing mutations is a reflection of the overlay of genetic diversity on primary disease phenotype. Numerous examples exist of the effects of these so-called disease-modifying genes producing phenotypic variations among individuals with the identical primary disease-causing mutations (gene-gene interactions) and the effects of disease-modifying genes interacting with environmental determinants to alter phenotype further (geneenvironment interactions). These interactions are clearly important in polygenic diseases; gene-gene and geneenvironment interactions can modify the phenotypic expression of the disease. Among patients with sickle cell disease, for example, some patients experience painful crises, whereas others exhibit acute chest syndrome; still other presentations include hemolytic crises. Genetic disorders affecting a unique pool of DNA, mitochondrial DNA, have been identified. Mitochondrial DNA is unique in that it is inherited only from the mother. In addition, mutations in mitochondrial DNA can vary among mitochondria within a given cell and within a given individual (heteroplasmy). Examples of genetic disorders based in the mitochondrial genome are Kearns-Sayre syndrome and Leber hereditary optic neuropathy. The list of known mitochondrial genomic disorders is growing rapidly, and mitochondrial contributions to a large number of common polygenic disorders may also exist.

MOLECULAR MEDICINE A principal goal of current molecular strategies is to restore normal gene function to individuals with genetic mutations. Methods to do so are currently primitive, and a number of obstacles must be surmounted for this approach to be successful. The principal problems are that to deliver a complete gene into a cell is not easy, and persistent expression of the new gene cannot be ensured because of the variability in its incorporation in the genome and the consequent variability in its regulated expression. Many approaches have been used to date, but none has been completely successful. They include the following: (1) packaging the cDNA in a viral vector, such as an attenuated adenovirus, and using the cell’s ability to take up the virus as a means for the cDNA to gain access to the cell; (2) delivering the cDNA by means of a calcium phosphate–induced perturbation of the cell membrane; and (3) encapsulating the cDNA in a liposome that can fuse with the cell membrane and thereby deliver the cDNA. After the cDNA has been successfully delivered to the cell of interest, the magnitude and durability of expression of the gene product are important variables. The magnitude of expression is determined by the number of copies of cDNA

Chapter 1—Molecular Basis of Human Disease taken up by a cell and the extent of their incorporation in the genome of the cell. The durability of expression appears to be dependent partly on the antigenicity of the sequence and protein product. Notwithstanding these technical limitations, gene therapy has been used to treat adenosine deaminase deficiency successfully, which suggests that the principle on which the treatment is based is reasonable. Clinical trials of gene therapy have slowed considerably after unexpected deaths were widely reported in both the scientific and lay media. Efforts on other genetic disorders and as a means to induce expression of a therapeutic protein (e.g., vascular endothelial cell growth factor to promote angiogenesis in ischemic tissue) are ongoing. Understanding the molecular basis of disease leads naturally to the identification of unique disease targets. Recent examples of this principle have led to the development of novel therapies for diseases that have been difficult to treat. Imatinib, a tyrosine kinase inhibitor that is particularly effective at blocking the action of the bcr-abl kinase, is effective for the treatment of chronic-phase chronic myelogenous leukemia. Monoclonal antibody to tumor necrosis factor-α (infliximab) and soluble tumor necrosis factor-α receptor (etanercept) are prime examples of biologic modifiers that are effective in the therapy of chronic inflammatory disorders, including inflammatory bowel disease and rheumatoid arthritis. This approach to molecular therapeutics is rapidly expanding and holds great promise for improving the therapeutic armamentarium for a variety of diseases. Beyond cancer-related categories (e.g., DNA, RNA repair), gene expression arrays have provided additional interactions of regulatory pathways of clinical interest. The limitation of gene expression profile using micro-arrays, which does not account for post-transcriptional and other post-translational modifications of protein-coding products, will likely be overcome by approaches and advances in proteomics. Such processes by signaling networks tend to amplify or attenuate gene expression on time scales lasting seconds to weeks. Much work remains to improve current knowledge about the pathways that initiate and promote tumors. The basic pathways and nodal points of regulation will be identified for rational drug design and target from mechanistic insights gleaned from expression profiling of cultured cell lines, from small animal models of human disease, and from human samples. Although accounting for tissue heterogeneity and variation among different cell types, the new systems’ approach for incorporating genomic and computational research appears particularly promising to decipher the pathways that promote tumorigenesis. In turn, biologists and clinicians will use information derived from these tools to understand the events that promote survival, proangiogenesis, and immune escape, all of which may confer metastatic potential and progression. What potential diagnostic tools are available to establish genetic determinants of drug response? Genome-wide approaches from the Human Genome Project in combination with micro-arrays, proteomic analysis, and bioinformatics will identify multiple genes encoding drug targets (e.g., receptors). Similar high-throughput screening should provide insights into the predisposition to adverse effects of outcomes from treatments that are linked to genetic polymorphisms.

PHARMOCOGENETICS The future of pharmacogenetics is to know all the factors that influence adverse drug effects. In this way, the premature abandonment of special drug classes can be avoided in favor of rational drug design and therapy. Many hurdles must be overcome for pharmacogenetics to become more widespread and to be integrated into medical practice. Current approaches of trial and error in medical practice are well engrained on the parts of physicians. In addition, the allure for blockbuster drugs by the pharmaceutical industry warrants a new model for approaching individualized doses. New training for physicians in molecular biology and genetics should complement clinical pharmacogenomic studies that determine efficacy in an era of evidence-based medicine. Pharmacogenetic polymorphisms, unlike other clinical variables such as renal function, need only a single test, ideally as a newborn. Polygenic models of therapeutic optimization still face hurdles that reduce the chances for abuse of genetic information and additional costs. On the other hand, SNP haplotyping has the potential to identify genetically similar subgroups of the population and to randomize therapies based on more robust genetic markers. On a population level, genomic variability is much greater within than among distinct racial and ethnic groups. Both therapeutic efficacy and host toxicity are influenced by the patient’s specific disease, age, renal function, nutritional status, and other co-morbid factors. New challenges will be posed for the selection and guide to drug therapy for patients with cancer, hypertension, and diabetes. It is conceivable that treatment of multisystem disorders (e.g., metabolic syndrome) might be derived from novel therapeutics based on individual, interacting, and complementary molecular pathways.

REGENERATIVE MEDICINE A new era of regenerative biology has emerged with the discovery that adult mammalian cells can be reprogrammed into new cells. Regenerative medicine entails novel applications and approaches to exploit the resident population of progenitor cells for regeneration or repair of damaged tissues. After irreversible damage, transplantation of solid organs such as the heart, kidney, and lungs is a wellestablished medical-surgical intervention, but the limited availability of organs restricts widespread applications. Manipulation of cultured cells for transplantation heralds an alternative and complementary strategy to solid organ transplantation and offers an expanded platform for regenerative medicine. Although postmitotic, terminally differentiated organs are devoid of significant regenerative capacity, recent evidence for cellular plasticity of adult solid organs, throughout adult life, has challenged this prevailing dogma. This makeover involves approaches either to convert adult into pluripotent stem cells—retaining the ability to differentiate into new cell types—or forced reprogramming of adult cells into mature or progenitor cells. Embryonic stem (ES) cells share common features of clonagenicity, self-renewal, and multi-potentiality, a prerequisite for differentiation into diverse cell lineages of multicellular adult organism. Both technical and ethical concerns propelled the search for new sources, including

Section I—Introduction to Molecular Medicine

the isolation of ES cells from a single blastomere, which circumvents destruction of the embryo, and the use of postimplantation embryos as ES cell donors. Somatic cell nuclear transplantation (SCNT) or nuclear transfer (NT) is a technique for successful cloning and reprogramming of adult animal cell nuclei from healthy oocyte host cells. SCNT provides a source of stem cells tailored to the donor organism and promises to accelerate the pace for human use. Induced pluripotent stem (iPS) cells share the common features of somatic cell reprogramming but with the aid of four transcription factors by retroviral transduction. Whether symmetrical and asymmetrical cell division promotes the differentiation of pluripotent progenitor cells into distinct lineages of the mature organ awaits future studies. It is conceivable that age, gender, risk factors, and other disease status will have an impact on regenerative

plasticity, proliferation, or cellular functions. Another future hurdle will be to determine whether genetic factors enhance the cellular and molecular properties of ES cells essential for the reconstitution of a well-differentiated organ in vivo. Might progenitor cells derived from bone marrow or circulating blood be administered safely and efficaciously? Both clinical and translational scientists are being asked to address whether stem cell therapy has efficacy for the current victims of either stroke or heart attack. Beyond the feasibility are questions related to benefits from transplantation of different cells originating from embryonic, fetal, or adult stem cell lineages. Whether priming of endogenous cell-mediated repair mechanisms using genetically engineered cell lines leads to improvement in selected endpoints and clinical outcomes awaits large-scale clinical trials.

Prospectus for the Future The concept of personalized medicine will be realized from the functional and analytical phenotyping that aids diagnosis and treatment based on the individual’s genome and disease profile. An important future challenge will be the extraction of biologically meaningful data of direct clinical relevance to diagnosis, prognosis, therapeutic response, and, ultimately, prevention. What are the functional consequences of genome occupancy and modification in health and disease? Computational analyses will play an increasing role in understanding cancer pathogenesis and the mechanisms of disease. Information about the hierarchy of cellular functions is being coupled with powerful approaches to derive different yet complementary perspectives about molecular mechanisms. Micro-array analysis has already

References Acharya MR, Sparreboom A, Venitz J, Figg WD: Rational development of histone deacetylase inhibitors as anticancer agents: A review. Mol Pharmacol 68:917-932, 2005. Collins FS, Green ED, Guttmacher AE: A vision for the future of genomics research. Nature 422:835-847, 2003. Evans WE, McLeod HL: Pharmacogenomics: Drug disposition, drug targets, and side effects. N Engl J Med 348:538-549, 2003.

provided new classes of hematologic diseases and prognostic factors in breast cancer. Experimental approaches are already underway to reduce tumorigenesis into discrete modules of regulatory networks and biologic processes. A catalog of listed genes that change with tumor type, for example, should not be equated with prognosis, therapeutic response, or adverse outcomes. How to move diagnostic tools using micro-arrays and gene expression profiles into clinical decision making will be the focus of research programs in translational and clinical outcomes. Specific therapies for many inheritable diseases have lagged substantially behind advances in other fields, but new opportunities appear on the horizon for improving prognosis and clinical outcomes in the era of regenerative medicine.

Hinds DA, Stuve LL, Nilsen GB, et al: Whole-genome patterns of common DNA variation in three human populations. Science 307:1072-1079, 2005. Krause DS, Van Etten RA: Tyrosine kinases as targets for cancer therapy. N Engl J Med 353:172-187, 2005. van Steensel B: Mapping of genetic and epigenetic regulatory networks using microarrays. Nat Genet 37:S18-S24, 2005. Willard HF, Ginsburg GS (eds): Genomic and Personalized Medicine. New York, Elsevier, 2009. Zamore PD, Haley B: Ribo-gnome: The big world of small RNAs. Science 309:1519-1524, 2005.

Section II Evidence-Based Medicine

Evidence-Based Medicine, Quality of Life, and the Cost of Medicine – TARIQ * BELAND

Chapter

Evidence-Based Medicine, Quality of Life, and the Cost of Medicine Sara G. Tariq and Susan S. Beland

he diagnosis and treatment of individual patients involve clinical experience and skills on the part of the physician and knowledge of scientific information obtained through clinical trials. In the past, most of the daily practice was based on informal learning and a tradition of knowledge transferred from experienced clinicians to trainees and colleagues. Increasingly, however, this informal technique is being supplanted by rigorous analysis of the scientific underpinnings of clinical logic. Electronic databases and Internet technology enable collation and dissemination of information to help identify which techniques are supported by clinical trials. Evidence-based medicine has evolved during the past decade and uses the best available evidence from published research as the foundation for clinical decision making. This foundation, in addition to clinical expertise and a respect for patient preference, will aid the physician in providing optimal outcomes and a quality of life for the patient. However, the development of new techniques in medicine, often at great cost, can strain the ability of a society to fund and provide such services. Critical appraisal of both new and traditional diagnostic and treatment modalities is thus needed.

Critical Appraisal of the Literature Being cognizant of the types of evidence is crucial to practice evidence-based medicine. Research studies can be divided into two major categories: primary and secondary (Table 2-1). Primary studies can have a number of designs. In ran domized controlled studies, participants in the trial are ran16

domly allocated to one intervention or another. Both groups are followed for a specified period and analyzed in terms of specific outcomes defined at the outset of the study. This type of study allows rigorous assessment of a single variable in a defined patient group, has a prospective design that potentially eradicates bias by comparing two otherwise similar groups, and allows for meta-analysis. However, these studies are expensive and time-consuming. Results of randomized controlled trials can have enormous impact on the practice of medicine, as exemplified by the Women’s Health Initiative randomized controlled trial. This study was designed to assess the risks and benefits for postmenopausal hormone use in healthy women. However, the trial was stopped early because of an increased incidence of breast cancer, coronary heart disease, stroke, and thromboembolic disease in the hormone-treated group. Cohort studies have two or more groups of participants selected on the basis of differences in their exposure to a particular agent. The participants are prospectively followed to see how many in each group develop a disease or other specific outcome. A well-known example is the Framingham Heart Study that enrolled 5200 participants in 1948 and followed them forward in time to examine the progression and risk factors for heart disease. The data provided from the Framingham Study have helped clinicians understand the development and progression of heart disease and its risk factors. As with randomized trials, cohort studies are time-consuming. Case-control studies involve patients with a particular disease or condition who are identified and matched with control patients. The control participants can be patients with another disease or individuals from the general population. The validity of these retrospective studies depends on careful selection of the control group. For example, the impact of risk factors for men and women was recently evaluated in the CARDIO 2000 Study. The authors

Chapter 2—Evidence-Based Medicine, Quality of Life, and the Cost of Medicine Table 2-1 Types of Research Studies

Table 2-2 Requirements of Screening Tests

Primary Studies

Secondary Studies

Randomized control Case control Cohort studies Cross sectional Case series Case report

Meta-analyses Clinical practice guidelines Decision analysis Cost-effectiveness analysis

• Prevalence of disease must be sufficiently high. • Disease must have significant morbidity and mortality rates. • Effective treatment must be available. • Improved outcomes from early diagnosis and treatment must be present. • Test should have good sensitivity and specificity parameters. • Test should carry acceptable risks and be cost-effective.

evaluated 848 hospitalized patients after their first episode of acute coronary syndrome and used 1078 age- and sexmatched controls. The data revealed that women experiencing their first event were significantly older than men. Case reports describe the medical history of a single patient. When medical histories of more than one patient with a particular condition are described together to illustrate one aspect of the disease process, the term case series is used. Secondary (integrative) studies attempt to summarize and draw conclusions from primary information. Meta-analyses use statistical techniques to combine and summarize the results of primary studies. By combining the results from many trials, meta-analyses are able to estimate the magnitude of the effect of an intervention or risk factor as well as evaluate previously unanswered questions by performing subgroup analyses. The use of meta-analysis has provoked some controversy. Some investigators believe that metaanalyses may be as reliable as randomized controlled trials, whereas others believe that the technique should be used only as an alternate to randomized trials. However, in the absence of a large randomized controlled study, a metaanalysis of multiple smaller studies may be the best source of information to answer a specific question. Clinical practice guidelines attempt to summarize diagnostic and treatment strategies for common clinical problems to assist the physician with specific circumstances. They are usually published by medical organizations, such as the American College of Physicians, and government agencies, such as the Agency for Health Care Policy and Research and the United States Preventive Services Task Force. Decision analysis uses the results of primary studies to generate probability trees to aid both health professionals and patients in making choices about clinical management. Cost-effectiveness analysis evaluates whether a particular course of action is an effective use of resources.

Testing in Medical Practice Screening tests are performed on asymptomatic healthy people to detect occult disease and should meet the criteria listed in Table 2-2. Screening tests are most useful when a high prevalence of disease is present in the population and the test has adequate sensitivity and specificity parameters. When applied to a disease with low prevalence, a test with low specificity would have an unacceptable number of falsepositive results, which would lead to further procedures that are often invasive and expensive.

Diagnostic tests are used to determine the cause of illness in symptomatic persons and can be helpful in patient management by evaluating the severity of disease, determining prognosis, detecting disease recurrence, or selecting appropriate medications or other therapies. When considering diagnostic tests, the physician should weigh the potential benefits against the risks and expense. When comparing the efficacy of a new diagnostic test, the critical issues are the following: (1) Does the new test have something to offer that the currently accepted test does not? (2) Does the new test provide additional information that alters the post-test probability, which is the likelihood that a patient who has a positive test has the disease? Comparing the post-test probability with the pre-test probability before ordering the test, which is the clinical assessment of diagnostic possibilities, is also important. Values for some pre-test probabilities have been published, but more often they are derived from the physician’s clinical experience and are influenced by the practice setting. For instance, an obese African-American woman from the rural South is experiencing fatigue, blurry vision, and frequent vaginal yeast infections, and she has a strong family history of diabetes. Based on these features, she would have a high pre-test probability for type 2 diabetes mellitus. If a new diagnostic test were available for the diagnosis of diabetes, a comparison could be made on the post-test probabilities expected from the standard test (fasting blood glucose) and the new test. Ideally, the new test would offer greater diagnostic accuracy. Sensitivity and specificity are important parameters to consider when evaluating a diagnostic test. Sensitivity is an index of the diagnostic test’s ability to detect the disease when it is present. Specificity is the ability of the diagnostic test to identify correctly the absence of the disease. These parameters are calculated by the use of a 2 × 2 table (Table 2-3). An additional value, the likelihood ratio, which uses

Table 2-3 Schematic Outcomes of a Diagnostic Test (2 × 2 Table) Test Result

Disease Present

Disease Absent

Positive Negative

True positive (a) False negative (c)

False positive (b) True negative (d)

Positive predictive value (true-positive rate) =a/(a+b). Negative predictive value (false-positive rate) =d/(c+d). Sensitivity =a/(a+c); patients with the disease who have a positive test. Specificity =d/(b+d); patients without the disease who have a negative test.

Section II—Evidence-Based Medicine

both sensitivity and specificity, gives an even better indication of the test’s performance. A high positive likelihood ratio indicates a high likelihood of the presence of disease, whereas a high negative likelihood ratio identifies the absence of disease. Positive likelihood ratio: Sensitivity ( probability that test is positive in diseased patients ) = 1 − Specificity ( probability that test is positive in nondiseased patients)

Negative likelihood ratio: 1 − Sensitivity ( probability that test is negative in diseased d patients ) = 1 − Specificity ( probability that test is negative in nondiseased patients) After determining the validity of the diagnostic test, its applicability to the patient in question and whether the test is affordable and accurate in a particular setting should be ascertained. If the diagnostic test requires special devices or skills that are not available in the practice facility, the results provided can be inaccurate. Most importantly, an assessment should be made about whether the test will change the management offered or decrease the need for the use of other tests.

Evaluating Evidence about Treatment One of the most common problems facing physicians is the need to assess the validity of new treatments being developed as well as validity of traditional treatments that have been used for years. For example, how long after discharge from the hospital should treatment with antimicrobial agents continue for the patient who had been hospitalized for pneumonia? What is the value of plasmapheresis in thrombotic thrombocytopenic purpura? The first step in evaluating prospective treatments is to assess whether the information is derived from a properly conducted randomized controlled study. Every patient who enters the trial must be accounted for at the end of the study. The patients who are lost to follow-up often have different outcomes. If the conclusion of the trial does not change after accounting for the lost patients, then validity is added to the study. Another point to consider is whether patients were analyzed in their original randomized groups even if they did not undergo the intervention in question. This is termed an intention-to-treat analysis. A description of whether both groups were treated differently regarding other interventions (e.g., co-interventions) should be included. Assessing the importance of the data provided is the next step. This includes a number of simple statistical calculations

applied to the available results. The first is relative risk reduction (RRR): Incidence of outcome in control group − Incidence of outcome in study group RRR = Incidence of outcome in control group For example, the Diabetes Control and Complications Trial investigated the effect of tight control of blood glucose in patients with type 1 diabetes on the development and progression of long-term complications. The study involved more than 1400 patients, with one half randomized to intensive treatment and one half to conventional therapy. In this study, 3.4% of the patients in the conventional group and 2.2% in the intensive group developed microalbuminuria, indicating a 35% decrease in the occurrence of microalbuminuria in the primary prevention group: RRR =

0.034 − 0.022 × 100 = 35% 0.034

The greater the RRR, the more effective the therapy. However, the RRR does not take into account the baseline risk of the patients entering the trial and thus does not differentiate between large and small effects. The significance of RRR is discussed on the web site (Web Text 2-1). Calculating the absolute risk reduction (ARR), which gives the absolute difference in rates between the two groups, is another way of assessing the outcome. The ARR is defined as the number (X) that had the ill effect in the control group minus the number (Y) in the treatment group (i.e., ARR = X − Y). Using the previous example, the ARR for the development of microalbuminuria is 0.034 − 0.022 = 0.012, or 1.2%. Another valuable calculation is the number needed to treat, which represents the number of patients who need to be treated to prevent a single outcome event and is the inverse of the ARR (i.e., 1/[X − Y]). The lower the number needed to treat, the more clinically relevant is the treatment. Again, using the example, to prevent 1 patient from developing microalbuminuria, 83 patients with diabetes would have to be treated with intensive therapy (1/[X − Y] = 1/0.012 = 83). From this example, what seems like a large RRR of 35% actually translates to a relatively small (although significant) number of patients who benefited from intensive treatment. As before, assessment of the applicability of this information to a particular patient should be made by taking into account whether the patient in question has the same characteristics of the patients included in the study. Evidence of side effects, cause, or value of a particular clinical sign in diagnosis can be assessed along these same lines.

Internet in Clinical Practice The use of computer systems for disseminating medical information has increased exponentially. Numerous worldwide websites offer high-quality medical news, information

Chapter 2—Evidence-Based Medicine, Quality of Life, and the Cost of Medicine Table 2-4 Worldwide Websites • Cochrane Collaboration—one of the major organizations involved in evidence-based medicine (http://www.cochrane. org) • MD Consult—comprehensive medical information service (http://www.mdconsult.com) • Centers for Disease Control and Prevention (http://www .cdc.gov) • National Institutes of Health (http://www.nih.gov) • UpToDate—comprehensive clinical information website that is constantly updated (http://www.uptodate.com) • Student Consult—provides access to full standard texts online (http://www.studentconsult.com)

about practice guidelines, online textbooks and journals, and information about evidence-based medicine. In addition, many government sites offer up-to-date information (e.g., Centers for Disease Control and Prevention, National Institutes of Health). Table 2-4 lists some of these sites.

Including the Patient in the Decision Process Searching for the best evidence and applying it have the ultimate goal of providing better patient care. The process should also involve informing the patient of the available options and offering options based on good evidence. Effective communication, geared toward the patient’s level of health literacy, is crucial to ensure that the patient makes an informed decision. Using a certain therapy or implementing a diagnostic test may be inconvenient, or the patient may develop a certain side effect that he or she is not willing to accept. Involvement of the patient in the decisionmaking process requires good communication and adequate resources for patient education.

Quality of Life Health care in the millennium has changed significantly. An increasing number of patients survive illnesses that used to be fatal, and many patients have multiple co-existing illnesses. Assessing clinical improvement to a given treatment covers only one aspect of the clinician’s success. For example, although survival is an important outcome for patients with cancer, overall quality of life is fundamental. A patient can have improvement in disease-free survival without having a significant change in quality of life, and vice versa. Quality of life represents a subjective concept that is defined by the subjective perception of the patient and includes physical, emotional, social, and cognitive functions and the disease symptoms and side effects of a given treatment or intervention. For example, in examining the efficacy of a drug for postchemotherapy anemia, it would not only be important to know whether the hemoglobin rises appropriately but also to know whether the patient subjectively has more energy and is able to perform the normal duties of life. Quality of life is more commonly becoming a defined outcome measure in clinical trials. An increasing number of studies have been conducted in which health-related quality

of life is either the primary or secondary endpoint. Hopefully, clinicians can then take the information gained from these data and apply it in a holistic manner to optimize patient care.

COST OF MEDICINE The practice of medicine has significantly changed during the past 30 years. The cost of medicine has risen astronomically, and it is the duty of the physician to be cognizant of this in daily practice. Health care spending is growing much faster than the rest of the economy. Rising hospital expenses reflect many factors, including the demand for new medications and technology as well as the aging population. Physicians can contribute to the reduction of costs by being aware of medication prices and ordering tests appropriately. The pharmaceutical industry has been accused of contributing to medical inflation. The industry spends more than $11 billion annually on promotion and marketing, and $8,000 to $13,000 per physician each year. They employ 1 drug representative for every 11 physicians in the United States. The average price of drugs rose almost 50% between 1992 and 2000. Literature suggests that the gifts and perks physicians receive from the pharmaceutical representatives have a major influence on their practices and prescribing habits. Caution is recommended when analyzing data from pharmaceutical representatives, taking into account the inherent bias that exists regarding the medication they are marketing. The medical profession is responsible for providing the best care possible for patients, and barriers to this care arise when a gift or amenity accepted from a pharmaceutical representative obscures the judgment of appropriate and cost-effective care. Generic drugs should be prescribed whenever possible. Studies have shown that if physicians substituted generic drugs for brand-name drugs, the potential national savings would be up to 5.9 billion dollars annually. In addition, all medical schools should stress the use of generic drugs to students and residents. The use of tests is the second area in which physicians must be prudent when it comes to cost. The routine ordering of expensive and unnecessary tests has become part of the medical culture, but they can never take the place of a thorough history and physical examination. Evidence-based medicine is an important tool to use when deciding whether a certain diagnostic test is needed to help in the care of a patient. The risks and benefits of each test that is ordered must be weighed against the costs. For example, asymptomatic patients who are concerned about ovarian cancer may want their physician to order a pelvic ultrasound. The prudent physician will know that the prevalence of ovarian cancer is low in the population and thus the literature does not support the routine use of pelvic ultrasound as a screening tool. Therefore, a pelvic ultrasound is not a costeffective test for screening ovarian cancer in all female patients. These are two ways in which physicians can take an active role in helping reduce the cost of medical care in this nation, but the problem is clearly larger than this. Physicians will need to find a balance between being cost-conscious and maintaining high-quality patient care as the medical field continues to expand.

Section II—Evidence-Based Medicine

Prospectus for the Future Challenges to be met: • Medical schools need to expand the teaching of evidencebased medicine to students and physicians in training. • Risks and benefits of screening tests need to be better defined (e.g., use of computed tomography in the diagnosis of early lung cancer versus risk for radiation exposure).

References Bhat SK: The cost of medicine. Ann Intern Med 139:74-75, 2003. Bottomley A: The cancer patient and quality of life. Oncologist 7:120-125, 2002. Diabetes Control and Complications Trial Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 329:977-986, 1993. Haas JS, Phillips KA, Gerstenberger EP: Potential savings from substituting generic drugs for brand-name drugs: Medical expenditure panel survey, 1997-2000. Ann Intern Med 142:891-897, 2005.

• To affect the cost of medicine, the overuse of technology (e.g., computed tomography for every patient with abdominal pain) needs to be addressed from the standpoint of evidence-based medicine. • Finally, universal health care coverage for all people in the United States is desperately needed.

Hall WJ: The ethical dilemma of accepting gifts from drug makers. ACP-ASIM Observer, December, 2001. Sacket D: Evidence-Based Medicine, 2nd ed. Oxford, Churchill Livingstone, 2000, pp 13-29. Writing Group for the Women’s Health Initiative Investigators: Risks and benefits of estrogen plus progestin in healthy postmenopausal women, JAMA 288:321333, 2002.

Section III Cardiovascular Disease

Structure and Function of the Normal Heart and Blood Vessels – MORSHEDZADEH * LI * BENJAMIN

Evaluation of the Patient with Cardiovascular Disease – LITWIN * BENJAMIN

Diagnostic Tests and Procedures in the Patient with Cardiovascular Disease – LITWIN

Heart Failure and Cardiomyopathy – LITWIN * BENJAMIN

Congenital Heart Disease –

Acquired Valvular Heart Disease –

Coronary Heart Disease –

Cardiac Arrhythmias –

Pericardial and Myocardial Disease –

WHITEHEAD LITWIN

MICHAELS

HAMDAN

STEHLIK * BENJAMIN

Other Cardiac Topics –

Vascular Diseases and Hypertension – VONGPATANASIN * VICTOR

BULL * BENJAMIN

III

Chapter

Structure and Function of the Normal Heart and Blood Vessels Jack Morshedzadeh, Dean Y. Li, and Ivor J. Benjamin

Gross Anatomy The heart is composed of four chambers, two atria and two ventricles, which form two separate pumps arranged side by side and in series (Fig. 3-1). The atria are low-pressure capacitance chambers that mainly function to store blood during ventricular contraction (systole) and then fill the ventricles with blood during ventricular relaxation (diastole). The two atria are separated by a thin interatrial septum. The ventricles are high-pressure chambers responsible for pumping blood through the lungs and to the peripheral tissues. Because the pressure generated by the left ventricle is greater than that generated by the right, the left ventricular myocardium is thicker than the right. The two ventricles are separated by the interventricular septum, which is a membranous structure at its superior aspect and a thick, muscular structure at its medial and distal portions. The atrioventricular (AV) valves separate the atria and ventricles. The mitral valve is a bileaflet valve that separates the left atrium and ventricle. The tricuspid valve is a trileaflet valve and separates the right atrium and ventricle. Strong chords (chordae tendineae) attach the ventricular aspects of these valves to the papillary muscles of their respective ventricles. These papillary muscles are extensions of normal myocardium that project into the ventricular cavities and are important for optimal valve closure. The semilunar valves separate the ventricles from the arterial chambers: the aortic valve separates the left ventricle from the aorta, and the pulmonic valve separates the right ventricle from the pulmonary artery. These valves do not have chordae. Rather, they are fibrous valves whose edges coapt closely, thus allowing for adequate valve closure. Each of the four valves is surrounded by a fibrous ring, or annulus, that forms part of the structural support of the heart. When open, the valves 22

allow free flow of blood across them and into the adjacent chamber or vessel. When closed, the valves effectively prevent the backflow of blood into the preceding chamber. The thin, double-layered pericardium surrounds the heart. The visceral pericardium is adherent to the heart and constitutes its outer surface, or epicardium. This outer surface is separated from the parietal pericardium by the pericardial space, which normally contains less than 50 mL of fluid. The parietal pericardium has attachments to the sternum, vertebral column, and diaphragm that serve to stabilize the heart in the chest. Normal pericardial fluid lubricates contact surfaces and limits direct tissue-surface contact during myocardial contraction. In addition, the normal pericardium modulates interventricular interactions during the cardiac cycle.

Circulatory Pathway The circulatory system is composed of two distinct and parallel vascular networks, arterial and venous networks, which interconnect through capillary beds of the distal target organs (see Fig. 3-1). Deoxygenated blood drains from peripheral tissues and enters the right atrium through the superior and inferior venae cavae. Blood draining from the heart enters the right atrium through the coronary sinus. This blood mixes in the right atrium during ventricular systole and then flows across the tricuspid valve and into the right ventricle during ventricular diastole. When the right ventricle contracts, blood is ejected across the pulmonic valve and into the main pulmonary artery, which then bifurcates into the left and right pulmonary arteries as these branches enter their respective lungs. After multiple bifurcations, blood flows into the pulmonary capillaries, where

Chapter 3—Structure and Function of the Normal Heart and Blood Vessels Head, upper extremities Superior vena cava Pulmonary artery

Pulmonary circulation

Bronchial arteries

Coronary circulation Ostium of coronary sinus

Aorta Left atrium

Right atrium Right ventricle Inferior vena cava

Left ventricle Abdominal viscera

“Capacitance” function of the venous system Venous valves

Lower extremities

“Resistance” function of the arterial system Figure 3-1 Schematic representation of the systemic and pulmonary circulatory systems. The venous system contains the greatest amount of blood at any one time and is highly distensible, accommodating a wide range of blood volumes (high capacitance). The arterial system is composed of the aorta, arteries, and arterioles. Arterioles are small muscular arteries that regulate blood pressure by changing tone (resistance).

Superior vena cava

carbon dioxide is exchanged for oxygen across the alveolarcapillary membrane. Oxygenated blood then drains from the lungs into the four pulmonary veins, which empty into the left atrium. During ventricular diastole, the blood flows across the open mitral valve and into the left ventricle. With ventricular contraction, the blood is ejected across the aortic valve and into the aorta and is subsequently delivered to the various organs, where oxygen and nutrients are exchanged for carbon dioxide and metabolic wastes. The heart receives blood through the left and right coronary arteries (Fig. 3-2). These are the first arterial branches of the aorta and originate in outpouchings of the aortic root called the sinuses of Valsalva. The left main coronary artery originates in the left sinus of Valsalva and is a short vessel that bifurcates into the left anterior descending (LAD) and the left circumflex (LCx) coronary arteries. The LAD travels across the surface of the heart in the anterior interventricular groove toward the cardiac apex. It supplies blood to the anterior and anterolateral left ventricle through its diagonal branches and to the anterior two thirds of the interventricular septum through its septal branches. The LCx traverses posteriorly in the left AV groove (between the left atrium and left ventricle), supplying blood to the lateral aspect of the left ventricle through obtuse marginal branches as well as giving off branches to the left atrium. The right coronary artery (RCA) originates in the right sinus of Valsalva and courses down the right AV groove to a point where the left and right AV grooves and the inferior interventricular groove meet, the crux of the heart. The RCA gives off atrial branches to the right atrium and acute marginal branches to the right ventricle. The blood supply to the diaphragmatic and posterior aspects of the left ventricle varies. In 85% of individuals, the RCA bifurcates at the crux into the posterior descending coronary artery (PDA), which travels in the inferior interventricular groove to supply blood to the inferior

Aortic arch

Superior vena cava

Pulmonary artery Left atrium

Left main coronary artery Left circumflex coronary artery Left anterior descending coronary artery (LAD) Diagonal branch of LAD Right ventricle Left ventricle Right coronary artery

Right coronary artery Figure 3-2 Major coronary arteries and their branches.

Section III—Cardiovascular Disease

left ventricular wall and inferior third of the interventricular septum and to the posterior left ventricular (PLV) branches, which supply the posterior left ventricle. This course is termed a right dominant circulation. In 10% of individuals, the RCA terminates before reaching the crux, and the LCx supplies the PLV and PDA. This course is termed a left dominant circulation. In the remaining individuals, the RCA gives rise to the PDA, and the LCx gives rise to the PLV in a co-dominant circulation. An understanding of coronary artery anatomy and distribution of blood supply enables the clinician to define the location of coronary artery disease based on history, physical examination, and noninvasive tests such as electrocardiography (ECG), echocardiography, and radionuclide ventriculography. Small vascular channels, called collateral vessels, interconnect the normal coronary arteries. These vessels are nonfunctional in the normal myocardium because no pressure gradient develops across them. However, in the setting of severe stenosis or complete occlusion of a coronary artery, the pressure in the vessel distal to the stenosis decreases, and a gradient develops across the collateral vasculature, resulting in flow through the collateral vessel. The development of collateral vasculature is directly related to the severity of the coronary stenosis and may be stimulated by ischemia, hypoxia, and a variety of growth factors. Over time, these vessels may reach up to 1 mm in luminal diameter and are almost indistinguishable from similarly sized, normal coronary arteries. Most of the venous drainage from the heart occurs through the coronary sinus, which runs in the AV groove and empties into the right atrium. A small amount of blood from the right side of the heart drains directly into the right atrium through the thebesian veins and small anterior myocardial veins.

Conduction System The electrical impulse that initiates cardiac contraction originates in the sinoatrial (SA) node, a collection of specialized pacemaker cells measuring 1 to 2 cm in length located high in the right atrium between the superior vena cava and the right atrial appendage (Fig. 3-3). The impulse then spreads through the atrial tissue and through preferential internodal tracts, ultimately reaching the AV node. This structure consists of a meshwork of cells located at the inferior aspect of the right atrium between the coronary sinus and the septal leaflet of the tricuspid valve. The AV node provides the only normal electrical connection between the atria and ventricles. After an electrical impulse enters the AV node, conduction transiently slows and then proceeds to the ventricles by means of the HisPurkinje system. The bundle of His extends from the AV node down the membranous interventricular septum to the muscular septum, where it divides into the left and right bundle branches. The right bundle branch is a discrete structure that extends along the interventricular septum and enters the moderator band on its way toward the anterolateral papillary muscle of the right ventricle. The left bundle branch is less distinct; it consists of an array of fibers organized into an anterior fascicle, which proceeds toward the anterolateral papillary muscle of the left ventricle, and a posterior fascicle, which proceeds posteriorly in the septum

Aorta Sinoatrial node

Bundle of His Main left bundle branch Atrioventricular node

Anterior fascicle of left bundle branch

Right bundle branch

Posterior fascicle of left bundle branch Figure 3-3 Schematic representation of the cardiac conduction system. Purkinje fibers

toward the posteromedial papillary muscle. Both the right and the left bundle branches terminate in Purkinje cells, which are large cells with well-developed intercellular connections that allow for the rapid propagation of electrical impulses. These impulse-generating cells then directly stimulate myocytes. Heart blocks, a form of cardiac arrhythmia, may arise from intrinsic problems of the conduction system or from impaired blood supply (coronary artery disease) to the conduction system. The SA node is supplied by the SA nodal artery, which is a branch of the RCA in about 60% of the population or a branch of the LCx in 40%. The AV node is supplied by the AV nodal artery, which is a branch of the RCA in about 90% of the population or a branch of the LCx in 10%. The right bundle branch receives most of its blood supply from septal perforators that branch off of the LAD. There may also be collateral blood supply from the RCA or LCx. The left anterior fascicle is supplied by septal perforators from the LAD and is particularly susceptible to ischemia and infarction. The proximal portion of the left posterior fascicle is supplied by the AV nodal artery and by septal perforators of the LAD. The distal portion of the posterior fascicle has a dual blood supply from anterior and posterior septal perforators (i.e., the LAD and PDA).

Neural Innervation The normal myocardium is richly innervated by the autonomic nervous system. The sympathetic supply is from preganglionic neurons located within the superior five to six thoracic segments of the spinal cord, which synapse with second-order neurons in the cervical sympathetic ganglia. Traveling within the cardiac nerves, these fibers end in the SA node, AV node, epicardial vessels, and myocardium. The parasympathetic supply is from preganglionic neurons originating in the dorsal motor nucleus of the medulla and pass as branches of the vagus nerve to the heart. Here the fibers synapse with second-order neurons located in ganglia within the heart. Nerve terminals of the parasympathetic nerves end in the SA node, AV node, epicardial vessels, and myocar-

Chapter 3—Structure and Function of the Normal Heart and Blood Vessels dium. A supply of vagal afferents from the inferior and posterior aspects of the ventricles mediate important cardiac reflexes, whereas the vagal efferent fibers to the SA and AV nodes are active in modulating impulse initiation and conduction. In general, sympathetic stimulation increases heart rate (HR) and force of myocardial contraction, and parasympathetic stimulation slows HR and reduces the force of contraction.

I band (actin)

A band (actin and myosin)

Z line

H zone (myosin)

Z line

Myocardium Cardiac tissue (myocardium) is composed of several cell types that together produce the organized contraction of the heart. Specialized myocardial cells make up the cardiac electrical system (conduction system) and are responsible for the generation of an electrical impulse and organized propagation of that impulse to cardiac muscle fibers (myocytes), which, in turn, respond by mechanical contraction. Atrial and ventricular myocytes are specialized, branching muscle cells connected end to end by intercalated disks. These thickened regions of the cell membrane (sarcolemma) aid in the transmission of mechanical tension between cells. The sarcolemma has functions similar to those of other cell membranes, including maintenance of ionic gradients, propagation of electrical impulses, and provision of receptors for neural and hormonal inputs. In addition, the sarcolemma is intimately involved with the coupling of myocardial excitation and contraction through small transverse tubules (T tubules) that extend from the sarcolemma into the intracellular space. The myocytes contain several other organelles: the nucleus; the multiple mitochondria responsible for generating the energy required for contraction; an extensive network of intracellular tubules called the sarcoplasmic reticulum, which functions as the major intracellular storage site for calcium; and the myofibrils, which are the contractile elements of the cell. Each myofibril is made up of repeating units called sarcomeres, which are, in turn, composed of overlapping thin actin filaments and thick myosin filaments and their regulatory proteins troponin and tropomyosin (Fig. 3-4).

Muscle Physiology and Contraction Contraction of myocytes begins with electrical depolarization of the sarcolemma, resulting in an influx of calcium into the cell through channels in the T tubules (Fig. 3-5). This initial calcium entry stimulates the rapid release of large amounts of calcium from the sarcoplasmic reticulum into the cell cytosol. The calcium then binds to the calciumbinding troponin subunit (troponin C) on the actin filaments of the sarcomere, resulting in a conformational change in the troponin-tropomyosin complex. This change facilitates the actin-myosin interaction, which results in cellular contraction. As the wave of depolarization passes, the calcium is rapidly and actively resequestered in the sarcoplasmic reticulum, where it is stored by various proteins, including calsequestrin, until the next wave of depolarization occurs. Calcium is also extruded from the cytosol by

A M line Sarcomere

Myosin

Actin

B Figure 3-4 A, Sarcomere as it appears under the electron microscope. B, Schematic of the location and interaction of actin and myosin.

various calcium pumps in the sarcolemma. The force of myocyte contraction can be regulated by the amount of free calcium released into the cell by the sarcoplasmic reticulum. More calcium allows for greater actin-myosin interaction, producing a stronger contraction. The energy for myocyte contraction is derived from adenosine triphosphate (ATP), which is generated by oxidative phosphorylation of adenosine diphosphate (ADP) in the abundant mitochondria of the cell. ATP is required both for calcium influx and for force generation by actin-myosin interaction. During contraction, ATP promotes dissociation of myosin from actin, thereby permitting the sliding of thick filaments past thin filaments as the sarcomere shortens. Under normal circumstances, fatty acids are the preferred energy source, although glucose can also be used as a substrate. These substrates must be constantly delivered to the heart through the bloodstream because minimal energy is stored in the heart. Myocardial metabolism is aerobic and thus requires a constant supply of oxygen. Under ischemic or hypoxemic conditions, glycolysis and lactate may serve as a source of ATP, although in insufficient quantities to sustain the working heart. Ischemic conditions might also promote alterations in both cytosolic and mitochondrial calcium overload, a major terminal event in muscle injury of the heart, termed myocardial infarction.

Section III—Cardiovascular Disease

Depolarization

Repolarization Na+-Ca2+ exchange pump

Ca2+

SR Ca2+ ATPase

Na+ Ca2+ release

Na+

Ca2+ storage

T tubule Ca2+

Ca2+

ATP Myosin

ATP

Ca2+-ATPase

Actin

Figure 3-5 Calcium dependence of myocardial contraction. (1) Electrical depolarization of the myocyte results in an influx of Ca2+ into the cell through channels in the T tubules. (2) This initial phase of calcium entry stimulates the release of large amounts of Ca2+ from the sarcoplasmic reticulum (SR). (3) The Ca2+ then binds to the troponin-tropomyosin complex on the actin filaments, resulting in a conformational change that facilitates the binding interaction between actin and myosin. In the presence of adenosine triphosphate (ATP), the actin-myosin association is cyclically dissociated as the thick and thin filaments slide past each other, resulting in contraction. (4) During repolarization, the Ca2+ is actively pumped out of the cytosol and sequestered in the SR. M, mitochondrion.

Circulatory Physiology and the Cardiac Cycle The cardiac cycle is a repeating series of contractile and valvular events during which the valves open and close in response to pressure gradients between different cardiac chambers (Fig. 3-6). This cycle can be divided into systole, the period of ventricular contraction, and diastole, the period of ventricular relaxation. With the onset of ventricular contraction, the pressure in the ventricles increases and exceeds that in the atria, at which time the AV valves close. Intraventricular pressure continues to rise, initially without a change in ventricular volume (isovolumic contraction), until the intraventricular pressures exceed the pressures in the aorta and pulmonary artery, at which time the semilunar valves open and ventricular ejection of blood occurs. With the onset of ventricular relaxation, the pressure in the ventricles falls until the pressure in the arterial chambers exceeds that

in the ventricles, and the semilunar valves close. Ventricular relaxation continues, initially without a change in ventricular volume (isovolumic relaxation). When the pressure in the ventricles falls below the pressure in the atria, the AV valves open, and a rapid phase of ventricular filling occurs as blood in the atria empties into the ventricles. At the end of diastole, active atrial contraction augments ventricular filling. This augmentation is particularly important in patients with poor ventricular function or stiff ventricles and is lost in patients with atrial fibrillation. In the absence of valvular disease, no impediment to the flow of blood exists from the ventricles to the arterial beds, and the systolic arterial pressure rises sharply to a peak. During diastole, the arterial pressure gradually falls as blood flows distally, and elastic recoil of the arteries occurs. This response contrasts with the pressure response in the ventricles during diastole, in which pressure gradually increases as blood enters the ventricles from the atria. Atrial pressure can be directly measured in the right atrium, whereas occluding

Chapter 3—Structure and Function of the Normal Heart and Blood Vessels Systole

Diastole

Table 3-1 Normal Values for Common Hemodynamic Parameters

120

Heart Rate Pressure (mm Hg)

60-100 Beats/Minute

Pressures

100 AVO

Aorta

80 60 40 20

Left ventricle

MVO v

LA a

4 1

Jugular pulse a

ECG P 0.0

0.1

0.2 0.3

Heart sounds

3 v

Systemic vascular resistance Pulmonary vascular resistance Cardiac output Cardiac index 0.5

0.6

≤9mmHg ≤9mmHg 15-30mmHg ≤9mmHg 15-30mmHg 3-12mmHg ≤12mmHg ≤12mmHg 100-140mmHg 3-12mmHg 100-140mmHg 60-90mmHg

Resistance

T 0.4

Central venous Right atrial Right ventricular Systolic End-diastolic Pulmonary arterial Systolic Diastolic Pulmonary capillary wedge Left atrial Left ventricular Systolic End-diastolic Aortic Systolic Diastolic

800-1500 dynes-sec/cm−5 30-120 dynes-sec/cm−5 4-6L/min 2.5-4L/min

0.7

Isovolumic contraction

Isovolumic relaxation Time (sec) Figure 3-6 Simultaneous electrocardiogram (ECG) and pressure tracings obtained from the left atrium (LA), left ventricle (LV), and aorta, and the jugular venous pressure during the cardiac cycle. For simplification, right-sided pressures have been omitted. Normal right atrial pressure closely parallels that of the left atrium, and right ventricular and pulmonary artery pressures are timed closely with their corresponding left-sided heart counterparts; they are reduced only in magnitude. The normal mitral and aortic valve closure precedes tricuspid and pulmonic valve closure, respectively, whereas valve opening reverses this order. The jugular venous pulse lags behind the right atrial pulse. During the course of one cardiac cycle, the electrical events (ECG) initiate and therefore precede the mechanical (pressure) events, and the latter precedes the auscultatory events (heart sounds) they themselves produce. Shortly after the P wave, the atria contract to produce the a wave. The QRS complex initiates ventricular systole, followed shortly by LV contraction and the rapid buildup of LV pressure. Almost immediately, LV pressure exceeds LA pressure, closing the mitral valve and producing the first heart sound. After a brief period of isovolumic contraction, LV pressure exceeds aortic pressure and the aortic valve opens (AVO). When the ventricular pressure once again falls below the aortic pressure, and the aortic valve closes to produce the second heart sound and terminate ventricular ejection. The LV pressure decreases during the period of isovolumic relaxation until it drops below LA pressure, and the mitral valve opens (MVO).

a small pulmonary artery branch and measuring the pressure distally (the pulmonary capillary wedge pressure) is often used to obtain left atrial pressure indirectly. An atrial pressure tracing is shown in Figure 3-6 and is composed of several waves. The a wave represents atrial contraction. As

the atria subsequently relax, the atrial pressure falls, and the x descent is noted on the pressure tracing. The x descent is interrupted by a small c wave, which is generated as the AV valve bulges toward the atrium during ventricular systole. As the atria fill from venous return, the v wave is seen, after which the y descent appears as the AV valves open and blood from the atria empties into the ventricles. The normal ranges of pressures in the various cardiac chambers are shown in Table 3-1.

CARDIAC PERFORMANCE The amount of blood ejected by the heart each minute is referred to as the cardiac output (CO) and is the product of the stroke volume (SV; amount of blood ejected with each ventricular contraction) and the HR: CO = SV × HR The cardiac index is the CO divided by the body surface area; it is measured in liters per minute per square meter and is a way of normalizing CO to body size. The normal CO at rest is 4 to 6 L/min, although this value can increase fourfold to sixfold during strenuous exercise as a result of increases in HR (chronotropic) and SV (inotropic). The SV is a measure of the mechanical function of the heart and is affected by preload, afterload, and contractility (Table 3-2). Preload is the volume of blood in the ventricle at the end of diastole and is primarily a reflection of venous return. Within limits, as the preload increases, the ventricle stretches, and the ensuing ventricular contraction becomes more rapid and forceful. This phenomenon is known as the

Section III—Cardiovascular Disease

Table 3-2 Factors Affecting Cardiac Performance Preload (left ventricular diastolic volume)

Afterload (impedance against which the left ventricle must eject blood)

Contractility (cardiac performance independent of preload or afterload)

Heart rate

Total blood volume Venous (sympathetic) tone Body position Intrathoracic and intrapericardial pressures Atrial contraction Pumping action of skeletal muscle Peripheral vascular resistance Left ventricular volume (preload, wall tension) Physical characteristics of the arterial tree (elasticity of vessels or presence of outflow obstruction) Sympathetic nerve impulses Increased contractility Circulating catecholamines Digitalis, calcium, other inotropic agents Increased heart rate or post-extrasystolic augmentation Anoxia, acidosis Decreased contractility Pharmacologic depression Loss of myocardium Intrinsic depression Autonomic nervous system Temperature, metabolic rate Medications, drugs

Frank-Starling relationship. Because ventricular volume is not easily measured, ventricular filling pressure (ventricular end-diastolic pressure, atrial pressure, or pulmonary capillary wedge pressure) is frequently used as a surrogate measure of preload. Two major means can manipulate the preload in a clinical setting. The first is to modulate volume status: intravenous fluids to increase preload and diuretics to decrease preload. The second is to regulate vascular tone: nitroglycerin to diminish preload. Afterload is the force against which the ventricles must contract to eject blood. The arterial pressure is often used as a practical measure of afterload; although, in truth, the intraventricular pressure, the size of the ventricular cavity, and the thickness of the ventricular walls (Laplace’s law) determine afterload. Thus, afterload is increased in the setting of systemic hypertension or stenosis of the aortic valve but may be equally increased in the setting of ventricular dilation or ventricular hypertrophy. Some antihypertensive drugs, such as angiotensin-converting enzyme (ACE) inhibitors and hydralazine, reduce blood pressure (BP) by reducing afterload. Contractility, or inotropy, although difficult to define, represents the force of ventricular contraction independent of loading conditions. For example, an increase in contractility results in a stronger ventricular contraction even when

the preload and afterload are kept constant. Direct stimulation from adrenergic nerves in the myocardium and circulating catecholamines released from the adrenal glands can alter contractility under normal conditions. Several medications have important positive inotropic effects that can be exploited clinically, including digoxin and the sympathomimetic amines (e.g., epinephrine, norepinephrine, dopamine). Other medications, many of them antihypertensive medications, (e.g., β blockers, calcium channel antagonists) have negative inotropic effects and can decrease the strength of ventricular contraction. Overall ventricular systolic function is frequently quantified by the ejection fraction, which is the ratio of the SV to the end-diastolic volume, that is, the fraction of blood in the ventricle ejected with each ventricular contraction. The normal ejection fraction is about 60% and can be measured by invasive (contrast ventriculography) or noninvasive (echocardiography or radionuclide ventriculography) methods. Clearly, systolic contraction is an important component of ventricular function; however, ventricular diastolic relaxation (lusitropy) also plays an important role in overall cardiac performance. Impaired relaxation (diastolic dysfunction), as occurs with ventricular hypertrophy or ischemia, results in a stiff, noncompliant ventricle, leading to impaired ventricular filling and an increased ventricular pressure for any given diastolic volume.

PHYSIOLOGY OF THE CORONARY CIRCULATION The heart is an aerobic organ requiring a constant supply of oxygen to maintain normal function. Under normal conditions, the supply of oxygen delivered to the heart is closely matched to the amount of oxygen required by the heart (the myocardial oxygen consumption [Mvo2]). The main determinants of Mvo2 are HR, contractility, and wall stress. The wall stress, as determined by Laplace’s law, is directly related to the systolic pressure and the heart size and inversely proportional to wall thickness: Wall stress = ( pressure × radius ) (2 × wall thickness ) Thus, the Mvo2 parallels changes in HR, BP, contractility, and heart size. In general, oxygen delivery to an organ can be augmented by either increasing blood flow or increasing oxygen extraction from the blood. For all practical purposes, the oxygen extraction by the heart is maximal at rest, and thus increases in coronary blood flow must meet increases in Mvo2. Because of the compression of intramyocardial blood vessels during systole, most coronary flow occurs during diastole. Therefore, diastolic pressure is the major pressure driving the coronary circulation. An important implication of this fact is that tachycardia, which primarily shortens the duration of diastole, results in reduced time for coronary flow, which occurs despite the increase in Mvo2 associated with increased HR. The systolic pressure has little effect on coronary blood flow except insofar as changes in BP lead to changes in Mvo2. Regulation of coronary blood flow occurs primarily through changes in coronary vascular resistance. In response

Chapter 3—Structure and Function of the Normal Heart and Blood Vessels to a change in Mvo2, the coronary arteries can dilate or constrict to allow for appropriate changes in coronary flow. Additionally, in the range of coronary perfusion pressures of 60 to 130 mm Hg, coronary blood flow is held constant by the process of autoregulation of the coronary arteries. This regulation of arterial resistance occurs at the level of the arterioles and is mediated by several factors. As ATP is metabolized during increased myocardial activity, adenosine is released and acts as a potent vasodilator. Decreased oxygen tension, increased carbon dioxide, acidosis, and hyperkalemia all develop during increased myocardial metabolism and may also mediate coronary vasodilation. The coronary arteries are innervated by the autonomic nervous system, and activation of sympathetic or parasympathetic neurons alters coronary blood flow by affecting changes in vascular tone. Parasympathetic innervation through the vagus nerve and through the neurotransmitter acetylcholine results in vasodilation. Sympathetic neurons use norepinephrine as a neurotransmitter and may have opposing effects on the coronary vasculature. Stimulation of α receptors results in vasoconstriction, whereas stimulation of β receptors leads to vasodilation. The ability of the coronary vasculature to mediate changes in blood flow through changes in vascular tone depends in large part on an intact, normally functioning endothelium. The endothelium produces several potent vasodilators, including endothelium-derived relaxing factor (EDRF) and prostacyclin. EDRF is likely to be nitric oxide or a compound containing nitric oxide and is released by the endothelium in response to acetylcholine, thrombin, ADP, serotonin, bradykinin, platelet aggregation, and an increase in shear stress. The latter stimulus accounts for the dilation of the coronary arteries in response to increases in blood flow in the setting of increases in Mvo2 (called flow-dependent vasodilation). Vasoconstricting factors, most notably endothelin, are also produced by the endothelium and are likely to play a role in regulating vascular tone. The balance of these vaso dilator and vasoconstriction factors may be important in conditions such as coronary vasospasm. In addition to influencing vascular tone, the endothelium has several other functions that have important implications for blood flow and tissue perfusion. These include maintenance of a nonthrombotic surface through inhibition of platelet activity, control of thrombosis and fibrinolysis, and modulation of the inflammatory response of the vasculature. Disturbances in these normal properties of the endothelium (endothelial dysfunction) are likely to play an important role in the pathophysiologic conditions of coronary atherosclerosis and thrombosis.

PHYSIOLOGY OF THE SYSTEMIC CIRCULATION The normal cardiovascular system is capable of providing appropriate blood flow to each of the organs and tissues of the body under a wide range of conditions. This is achieved by maintaining arterial BP within normal limits to meet functional needs from adjustments to the cardiac output and the resistance to blood flow in specific organs and tissues. Arterial pressure is regulated acutely and chronically through various local and systemic, humoral, and neural factors.

Poiseuille’s law describes the relationship between pressure and flow. Although not exactly descriptive of blood flow through elastic tapering blood vessels, Poiseuille’s law is useful in understanding blood flow. Fluid flow (F) through a tube is proportional (proportionality constant = K) to the pressure (P) difference between the ends of the tube: F = K × ∆P The reciprocal of K is the resistance to flow (R); that is K = 1/R. When fluid flows through a tube, the resistance to flow is determined by the properties of both the fluid and the tube. In the case of a steady, streamlined flow of fluid through a rigid tube, Poiseuille found that these factors determine resistance: R = 8ηL πr 4 where r is the radius of the tube, L is its length, and η is the viscosity of the fluid. This equation shows that the resistance to blood flow increases proportionately with increases in fluid viscosity or tube length. In contrast, radius changes have a much greater influence because resistance is inversely proportional to the fourth power of the radius. Poiseuille’s law incorporates the factors influencing flow, so that: F = ∆P R = ∆Pπr 4 8ηL The most important determinants of blood flow in the cardiovascular system are ΔP and r4. Thus, small changes in arterial radius can cause large changes in flow to a tissue or organ. Systemic vascular resistance (SVR) is the total resistance of flow offered by the blood vessels of the systemic circulation. Physiologic changes in SVR are primarily caused by changes in the radius of small arteries and arterioles, the resistance vessels of the systemic circulation. The SVR is defined as the pressure drop across the peripheral capillary beds divided by the blood flow across the beds (e.g., SVR = BP/CO). In practice, this is calculated as the mean arterial pressure minus the right atrial pressure divided by the cardiac output and is normally in the range of 800 to 1500 dynes sec/cm5. As with the coronary circulation autonomic innervation alters systemic vascular tone through sympathetic and parasympathetic innervation. Local oxygen tension, carbon dioxide levels, pH, and potassium levels have direct effects on vascular tone and blood flow. And finally, a normally functioning endothelium mediates changes in blood flow through potent vasodilatory and vasoconstricting factors (see “Physiology of the Coronary Circulation”). Control of BP through neural regulation occurs by means of tonic and reflexive modulation of autonomic nervous system outflow. Acutely, changes in these outflows influence key determinants of BP, such as cardiac chronotropy, inotropy, and vascular resistance. The primary mechanism by which BP is neurally modulated is through the baroreflexes. The baroreflex loop anatomically originates at the level of the baroreceptor. The baroreceptors are highly specialized stretch-sensitive nerve endings distributed throughout various regions of the cardiovascular systems, such as the

Section III—Cardiovascular Disease

carotid artery, aorta, and the cardiopulmonary region. Baroreceptors located in the carotid artery (e.g., carotid sinus) and aorta are sometimes referred to as high-pressure baroreceptors and those in the cardiopulmonary areas as lowpressure baroreceptors. After transmission of afferent impulses to the central nervous system, signals are integrated, and the efferent arm of the reflex projects neural signals systemically through the sympathetic and parasympathetic branches of the autonomic nervous system. In general, in response to an increase in systemic BP, there is an increased firing rate of the baroreceptors, efferent sympathetic outflow is inhibited (reducing vascular tone, chronotropy, and inotropy), and parasympathetic outflow is increased (reducing cardiac chronotropy). The opposite occurs when BP decreases. Cardiac output and systemic BP are controlled not only neurally and by local vasoactive substances through regulation of vascular tone but also by blood volume. A major physiologic control mechanism, which regulates total blood volume, is the renin-angiotensin-aldosterone system (RAAS). Renin is an enzyme secreted by the kidneys in response to low renal perfusion, low blood volume, low BP, or low sodium concentration. Renin converts the polypeptide angiotensinogen to angiotensin I in the liver. Angiotensin I circulates in the bloodstream and is converted to angiotensin II by the activity of ACE located primarily in the capillary beds of the lung. Angiotensin II is a powerful vasoconstrictor regulating BP through changes in vascular tone. In addition to its vasoactive properties, angiotensin II activates release of the hormone aldosterone from the adrenal cortex. Aldosterone then acts on the kidneys to retain sodium and thus water. Angiotensin II also acts directly on the posterior pituitary gland, increasing the secretion of vasopressin (e.g., antidiuretic hormone). Much like angiotensin II, vasopressin is a vasoconstrictor; it also acts on the kidney by retaining water through its action on V2 receptors in the collecting ducts. Although activity of the RAAS is to maintain blood volume, this system has adverse effects in chronic disease states, aggravating conditions such as hypertension and heart failure. Blood leaves the arterioles and flows into the capillary systems, where oxygen and nutrients are delivered to cells and carbon dioxide and metabolic wastes are removed. The deoxygenated blood then drains into peripheral veins, which contain valves to prevent backflow. These veins have thinner walls than arteries and function as capacitance vessels; they are able to accommodate a significantly larger volume of blood than the arterial system. With the aid of the pumping action of skeletal muscles and the respiratory motion of the chest wall, blood returns to the right atrium. This venous return can be altered by constriction or dilation of the peripheral veins. In addition to the venous drainage, a rich system of lymphatic vessels helps drain excess interstitial fluid from the periphery. The various lymphatic vessels drain into the thoracic duct and, subsequently, into the left brachiocephalic vein.

PHYSIOLOGY OF THE PULMONARY CIRCULATION Similar to the systemic circulation, the pulmonary circulation consists of a branching network of progressively smaller arteries, arterioles, capillaries, and veins. The pulmonary

capillaries are separated from the alveoli by a thin alveolarcapillary membrane through which gas exchange occurs. Carbon dioxide thus diffuses from the capillary blood into the alveoli, and oxygen diffuses from the alveoli into the blood. The flow of blood to various lung segments is regulated by several factors, the most important being the Po2 in the alveoli. In this manner, blood is shunted toward wellventilated lung segments and away from poorly ventilated segments. As a result of the extensive nature of the pulmonary capillary system and the distensibility of the pulmonary vasculature, the resistance across the pulmonary system (the pulmonary vascular resistance) is about 10% that of the systemic circulation. Owing to these features, the pulmonary system is able to tolerate significant increases in blood flow with little or no rise in pulmonary pressure. Thus, intracardiac shunts (e.g., atrial septal defects) may be associated with normal pulmonary pressure. The lung receives a dual blood supply. The pulmonary artery accounts for most pulmonary blood flow; however, the lungs also receive oxygenated blood through the bronchial arteries. These vessels supply oxygen to the lung and drain into the bronchial veins. The bronchial veins drain partly into the pulmonary veins; thus, a small amount of deoxygenated blood normally enters the systemic circulation and accounts for a physiologic right-to-left shunt. In the normal setting, this shunt is insignificant, accounting for only 1% of the total systemic blood flow.

Cardiovascular Response to Exercise The response of the heart to exercise is multifaceted and involves many of the previously discussed mechanisms of circulatory control (Table 3-3). In anticipation of exercise, neural centers in the brain stimulate vagal withdrawal and an increase in sympathetic tone, resulting in an increase

Table 3-3 Physiologic Responses to Exercise Increased heart rate Increased stroke volume Increased contractility Increased venous return

Decreased afterload Increased blood pressure

Increased O2 extraction

Increased sympathetic stimulation Decreased parasympathetic stimulation Increased sympathetic stimulation Sympathetic-mediated venoconstriction Pumping action of skeletal muscles Decreased intrathoracic pressure with deep inspirations Arteriolar vasodilation in exercising muscle Arteriolar vasodilation in exercising muscle (mediated chiefly by local metabolites) Increased cardiac output Vasoconstriction (sympathetic stimulation) on nonexercising vascular beds Shift in oxyhemoglobin dissociation curve as a result of local acidosis

Chapter 3—Structure and Function of the Normal Heart and Blood Vessels in HR and contractility (thus an increase in CO) before exercise ever starts. With exercise, sympathetic venoconstriction, augmented pumping action of skeletal muscles, and increased respiratory movements of the chest wall result in an increase in venous return to the heart. Through the Frank-Starling relationship, this increase in venous return results in an increase in contractility, thus augmenting CO. Sympathetic activation may also increase contractility; however, most of the increase in CO during exercise (up to 4 to 6 times the normal rate) is a consequence of an increase in HR. The peak HR that can be achieved is dependent on age and can be estimated by the following formula: maximal HR = (220 − age) ± 15 beats/minute. Local factors in exercising muscle cause arteriolar dilation, resulting in increased flow to the capillary beds. This vasodilation results in decreased resistance to flow, and

therefore the SVR decreases with exercise. Despite this change in resistance, the systolic BP rises, owing to the augmented CO and to sympathetic vasoconstriction, which leads to the preferential shunting of blood away from nonexercising vascular beds. The diastolic BP, by contrast, generally remains constant during exercise. The pulmonary system is able to tolerate the increased flow with only small increases in pulmonary pressure. The increases in HR and contractility result in a significant increase in Mvo2 (up to 300%), and coronary blood flow subsequently increases. Various types of exercises have different effects on the circulatory system. The response described in this text occurs with isotonic exercises, such as running or biking. With isometric exercises, such as weight lifting, the predominant response is an increase in BP, owing to an increase in peripheral vasoconstriction.

Prospectus for the Future Recent years have witnessed an explosion in the growth of basic knowledge governing normal heart development and function of the circulatory system. New insights about the molecular switches and factors that promote the formation of heart chambers and blood vessels are unraveling the genetic basis for unusual causes of congenital heart disease. Similarly, the recent discovery of specific growth factors and

References Berne RM, Levy MN: Physiology, Updated Edition, 5th ed., with Student Consult Access. Part IV: The Cardiovascular System. St Louis, Elsevier, 2004. Guyton AC, Hall JE: Textbook of Medical Physiology. St. Louis, Elsevier, 2005.

signaling pathways that guide vascular trajectory and ensure vascular stability offer new opportunities to rethink future strategies for promoting cardiac and vascular regeneration and repair. With increasing refinement in this basic knowledge, future milestones appear on the horizon for diagnosis, early treatment, and even prevention of cardiovascular diseases.

III

Chapter

Evaluation of the Patient with Cardiovascular Disease Sheldon E. Litwin and Ivor J. Benjamin

History As with diseases of most organ systems, the ability of the physician to diagnose diseases of the cardiovascular system is in large part dependent on eliciting and interpreting the patient’s clinical history. A thorough history can enable the physician to identify a patient’s symptoms as characteristic of a specific cardiovascular disorder or to suggest that symptoms are unlikely to be caused by cardiovascular disease. In addition, a complete history will reveal the presence of other systemic diseases that may have cardiovascular manifestations, identify existing risk factors that may be modified to prevent the future development of cardiovascular disease (see Chapter 9), enable the selection of appropriate further diagnostic testing (see Chapter 5), and allow the assessment of functional capacity and extent of cardiovascular disability. The patient should be asked about prior medical conditions, including childhood illnesses (e.g., rheumatic fever), as well as intravenous drug use, which may lead to the development of valvular heart disease. Several cardiovascular disorders are inherited (e.g., hypertrophic cardiomyopathy, Marfan syndrome, long QT syndrome), and a thorough family history may bring this potential to the examiner’s attention. The classic symptoms of cardiac disease include precordial discomfort or pain, dyspnea, palpitations, syncope or presyncope, and edema. Although characteristic of heart disease, these symptoms are nonspecific and may also occur as a result of diseases of other organ systems (e.g., musculoskeletal, pulmonary, renal, gastrointestinal). Furthermore, some patients with established cardiovascular disease may be asymptomatic or have atypical symptoms. Chest pain is a frequent symptom and may be a manifestation of cardiovascular or noncardiovascular disease (Tables 4-1 and 4-2). Full characterization of the pain with regard 32

to quality, quantity, frequency, location, duration, radiation, aggravating or alleviating factors, and associated symptoms may help distinguish among various causes. Reversible myocardial ischemia caused by obstructive coronary artery disease commonly results in episodic chest pain or discomfort during exertion or stress (angina pectoris). Patients frequently deny having pain and, instead, describe a discomfort in their chest. Sometimes they will refer to the discomfort as a squeezing, tightening, pressing, or burning sensation or as a heavy weight on their chest, and they will sometimes clench their fist over their chest while describing the discomfort (Levine sign). Anginal discomfort is classically located substernally or over the left chest. It frequently radiates to the epigastrium, neck, jaw, or back and down the ulnar aspect of the left arm. Radiation to the right chest or arm is less common, whereas radiation above the jaw or below the epigastrium is not typical of cardiac disease. Angina is usually brought on by either physical or emotional stress, is mild to moderate in intensity, lasts 2 to 10 minutes, and resolves with rest or sublingual administration of nitroglycerin. It may occur more frequently in the morning, in cold weather, after a large meal, or after exposure to environmental factors, including cigarette smoke, and is frequently accompanied by other symptoms, such as dyspnea, diaphoresis, nausea, palpitations, or lightheadedness. Patients frequently report a stable pattern of angina that is predictably reproducible with a given amount of exertion. Unstable angina occurs when a patient reports a significant increase in the frequency or severity of angina or when angina occurs with progressively decreasing exertion or at rest. When anginal-type pain occurs mainly at rest, it may be of a noncardiac origin, or it may reflect true cardiac ischemia resulting from coronary spasm (Prinzmetal or variant angina). The pain of an acute myocardial infarction may be similar

Chapter 4—Evaluation of the Patient with Cardiovascular Disease

Table 4-1 Cardiovascular Causes of Chest Pain Aggravating or Alleviating Factors

Condition

Location

Quality

Duration

Angina

Retrosternal region: radiates to or occasionally isolated to neck, jaw, shoulders, arms (usually left), or epigastrium

Pressure, squeezing, tightness, heaviness, burning, indigestion

30 minutes). The pain of acute pericarditis is usually sharper than anginal pain, is located to the left of the sternum, and may radiate to the neck or left shoulder. In contrast to angina, the pain may last hours, typically worsens with inspiration, and improves when the patient sits up and leans forward; it may be associated with a pericardial friction rub. Acute aortic dissection produces severe, sharp, tearing pain that radiates to the back and may be associated with asymmetrical pulses and a murmur of aortic insufficiency. Pulmonary emboli may produce the sudden onset of sharp chest pain that is worse on inspiration, is associated with shortness of breath, and may have an associated pleural friction rub, especially if a pulmonary infarction is present. A multitude of noncardiac conditions may also produce chest pain (see Table 4-2). The clinical history and physical examination findings will often help distinguish these causes from ischemic chest pain. Dyspnea, an uncomfortable, heightened awareness of breathing, is commonly a symptom of cardiac disease. Patients with decreased left ventricular function may exhibit significant abnormalities of the aortic or mitral valves or decreased myocardial compliance (i.e., left ventricular hypertrophy, acute ischemia), left ventricular diastolic, or left atrial pressure increases transmitted through the pulmonary veins to the pulmonary capillary system, producing vascular congestion. This congestion results in exudation of fluid into the alveolar space and impairs gas exchange across the alveolar-capillary membrane, producing the subjective

Precipitated by exertion, cold weather, or emotional stress; relieved by rest or nitroglycerin; variant (Prinzmetal) angina may be unrelated to exertion, often early in the morning Unrelieved by rest or nitroglycerin

Associated Symptoms or Signs Dyspnea; S3, S4, or murmur of papillary dysfunction during pain

Dyspnea, nausea, vomiting, weakness, diaphoresis Pericardial friction rub

Murmur of aortic insufficiency; pulse or blood pressure asymmetry; neurologic deficit

sensation of dyspnea. Dyspnea frequently occurs on exertion; however, in patients with severe cardiac disease, it may be present at rest. Patients with heart failure commonly sleep on two or more pillows because the augmented venous return that occurs on assuming the recumbent position produces an increase in dyspnea (orthopnea). In addition, these patients report awakening 2 to 4 hours after the onset of sleep with dyspnea (paroxysmal nocturnal dyspnea), which is likely caused by the central redistribution of peripheral edema in the supine position. Dyspnea may be associated with diseases of the lungs or chest wall and is also seen in anemia, obesity, deconditioning, and anxiety disorders. In addition, the sudden onset of dyspnea, with or without chest pain, may be present with pulmonary emboli. It is frequently difficult to distinguish cardiac from pulmonary causes of dyspnea by history alone because both may produce resting or exertional dyspnea, orthopnea, or cough. Wheezing and hemoptysis are classically results of pulmonary disease, although they are also frequently present in the patient with pulmonary edema resulting from left ventricular dysfunction or mitral stenosis. True paroxysmal nocturnal dyspnea is, however, more specific for cardiac disease. In patients with coronary artery disease, dyspnea may be an anginal equivalent; that is, the dyspnea is the result of ischemia and occurs in a pattern consistent with angina but in the absence of chest discomfort. Palpitation refers to the subjective sensation of the heart beating. Patients may describe a fluttering or pounding in the chest or a feeling that their heart races or skips a beat. Some

Section III—Cardiovascular Disease

Table 4-2 Noncardiac Causes of Chest Pain Aggravating or Alleviating Factors

Condition

Location

Quality

Duration

Pulmonary embolism (chest pain often not present)

Substernal or over region of pulmonary infarction

Pleuritic (with pulmonary infarction) or angina-like

Sudden onset (minutes to hours)

Aggravated by deep breathing

Pulmonary hypertension

Substernal

Pressure; oppressive

—

Aggravated by effort

Pneumonia with pleurisy

Located over involved area

Pleuritic

—

Aggravated by breathing

Spontaneous pneumothorax

Unilateral

Sharp, well localized

Sudden onset; lasts many hours

Aggravated by breathing

Musculoskeletal disorders

Variable

Aching, well localized

Variable

Herpes zoster

Dermatomal distribution Substernal or epigastric; may radiate to neck

Sharp, burning

Prolonged

Aggravated by movement; history of exertion or injury None

Burning, visceral discomfort

10-60min

Peptic ulcer

Epigastric, substernal

Prolonged

Gallbladder disease

Right upper quadrant; epigastric Often localized over precordium

Visceral burning, aching Visceral

Prolonged

Spontaneous or following meals

Variable; location often moves from place to place

Varies; often fleeting

Situational

Esophageal reflux

Anxiety states

people feel post-extrasystolic beats as a painful or uncomfortable sensation. Common arrhythmic causes of palpitations include premature atrial or ventricular contractions, supraventricular tachycardia, ventricular tachycardia, and sinus tachycardia. Occasionally, patients report palpitations even when no rhythm disturbance is noted during monitoring, as occurs commonly in patients with anxiety disorders. The pattern of palpitations, especially when correlated to the pulse, may help narrow the differential diagnosis: rapid, regular palpitations are noted with supraventricular tachycardia or ventricular tachycardia; rapid, irregular palpita-

Aggravated by large meal, postprandial recumbency; relief with antacid Relief with food, antacid

Associated Symptoms or Signs Dyspnea, tachypnea, tachycardia; hypotension, signs of acute right ventricular heart failure, and pulmonary hypertension with large emboli; pleural rub; hemoptysis with pulmonary infarction Pain usually associated with dyspnea; signs of pulmonary hypertension Dyspnea, cough, fever, bronchial breath sounds, rhonchi, egophony, dullness to percussion, occasional pleural rub Dyspnea; hyperresonance and decreased breath and voice sounds over involved lung Tender to palpation or with light pressure

Vesicular rash appears in area of discomfort Water brash

— Right upper quadrant tenderness may be present Sighing respirations; often chest wall tenderness

tions are noted with atrial fibrillation; and skipped beats are noted with premature atrial or ventricular contractions. Syncope is the transient loss of consciousness resulting from inadequate cerebral blood flow and may be the result of a variety of cardiovascular diseases (see Chapter 10). True syncope must be distinguished from primary neurologic causes of loss of consciousness (i.e., seizures) and metabolic causes of loss of consciousness (e.g., hypoglycemia, hyperventilation). Cardiac syncope occurs after an abrupt decrease in cardiac output, as may occur with acute myocardial ischemia, valvular heart disease (aortic or mitral stenosis),

Chapter 4—Evaluation of the Patient with Cardiovascular Disease hypertrophic obstructive cardiomyopathy, left atrial tumors, tachyarrhythmias (ventricular, or less commonly supraventricular, tachycardias), or bradyarrhythmias (e.g., sinus arrest, atrioventricular block, Stokes-Adams attacks). Reflex vasodilation or bradycardia may also result in syncope (vasovagal syncope, carotid sinus syncope, micturition syncope, cough syncope, or neurocardiogenic syncope), as may acute pulmonary embolism and hypovolemia. Because global, or at the very least bilateral, cortical ischemia is required to produce syncope, it rarely occurs as a result of unilateral carotid artery disease. However, syncope is occasionally the result of bilateral carotid artery disease and can also occur when disease of the vertebrobasilar system results in brainstem ischemia. In up to 50% of patients, the cause of a syncopal episode cannot be determined; however, in the cases in which a cause is determined, the most important factor in establishing the diagnosis is obtaining an accurate history of the event. Edema is a nonspecific symptom that commonly accompanies cardiac disease as well as renal disease (e.g., nephrotic syndrome), hepatic disease (e.g., cirrhosis), and local venous abnormalities (e.g., thrombophlebitis, chronic venous stasis). When edema occurs as a result of cardiac disease, it reflects an increase in venous pressure. This increased pressure alters the balance between the venous hydrostatic and oncotic forces, resulting in extravasation of fluid into the extravascular space. When this process occurs as a result of elevated left-sided heart pressure, pulmonary edema results, whereas elevated right-sided heart pressure results in peripheral edema. Characteristically, the peripheral edema of heart failure is pitting; that is, an indentation is left in the skin after pressure is applied to the edematous region. The edema is exacerbated by long periods of standing, is worse in the evening, improves after lying down, and may first be noted when a patient has difficulty in fitting into his or her shoes. The edema may shift to the sacral region after a patient lies down for several hours. When visible edema is noted, it is usually preceded by a moderate weight gain (i.e., 5 to 10lb), indicative of volume retention. As heart failure progresses, the edema may extend to the thighs and involve the genitalia and abdominal wall, and fluid may collect in the abdominal (ascites) or thoracic (pleural effusion) cavities. Anasarca with ascites should raise suspicion for constrictive peri carditis because this disease may progress very slowly and insidiously. Cyanosis is an abnormal bluish discoloration of the skin resulting from an increase in the level of reduced hemoglobin in the blood and, in general, reflects an arterial oxygen saturation of 85% or less (normal arterial oxygen saturation, ≥95%). Central cyanosis exhibits as cyanosis of the lips or trunk and often reflects right-to-left shunting of blood caused by structural cardiac abnormalities (e.g., atrial or ventricular septal defects) or pulmonary parenchymal or vascular disease (e.g., chronic obstructive pulmonary disease, pulmonary embolism, pulmonary arteriovenous fistula). Peripheral cyanosis may occur because of systemic vasoconstriction in the setting of poor cardiac output or may be a localized phenomenon resulting from venous or arterial occlusive or vasospastic disease (e.g., venous or arterial thrombosis, arterial embolic disease, Raynaud disease). When cyanosis occurs in childhood, it usually reflects congenital heart disease with right-to-left shunting of blood.

Myriad other symptoms, many of them nonspecific, may occur with cardiac disease. Fatigue frequently occurs in the setting of poor cardiac output or may occur secondary to the medical therapy of cardiac disease from overdiuresis, aggressive blood pressure lowering, or use of β-blocking agents. Nausea and vomiting frequently occur during an acute myocardial infarction and may also reflect intestinal edema in the setting of right ventricular heart failure. Anorexia and cachexia may occur in severe heart failure. Positional fluid shifts may result in polyuria and nocturia in patients with edema. In addition, epistaxis, hoarseness, hiccups, fever, and chills may reflect underlying cardiovascular disease. Many patients with significant cardiac disease are asymptomatic. Patients with coronary artery disease frequently have periods of asymptomatic ischemia that can be documented with ambulatory electrocardiographic monitoring. Furthermore, nearly one third of patients who suffer an acute myocardial infarction are unaware of the event. This silent ischemia appears to be more common in older adults and in patients with diabetes. Patients may also be asymptomatic despite having severely depressed ventricular function; this usually bespeaks a chronic, slowly progressive process. Reduced exercise capacity may only be seen during provocative testing. Similarly, recent findings show that a high percentage of episodes of atrial fibrillation are unrecognized by patients.

Assessment of Functional Capacity In patients with cardiac disorders, the ability or inability to perform various activities (functional status) plays an important role in determining their extent of disability, deciding when to institute various therapies or interventions, and assessing their response to therapy as well as determining their overall prognosis. The New York Heart Association Functional Classification is a standardized method for the assessment of functional status (Table 4-3) and relates functional capacity to the presence or absence of cardiac symptoms during the performance of usual activities. The

Table 4-3 Classification of Functional Status Class I

Uncompromised

Class II

Slightly compromised

Class III

Moderately compromised

Class IV

Severely compromised

Ordinary activity does not cause symptoms.* Symptoms only occur with strenuous or prolonged activity. Ordinary physical activity results in symptoms; no symptoms at rest. Less than ordinary activity results in symptoms; no symptoms at rest. Any activity results in symptoms; symptoms may be present at rest.

*Symptoms refer to undue fatigue, dyspnea, palpitations, or angina in the New York Heart Association classification and refer specifically to angina in the Canadian Cardiovascular Society classification.

Section III—Cardiovascular Disease

Physical Examination

merge with the c wave (c-v wave), thus diminishing or eliminating the x descent altogether. The y descent is attenuated in tricuspid stenosis, owing to the impaired atrial emptying. In pericardial constriction and restrictive cardiomyopathy, as well as in right ventricular infarction, the y descent becomes rapid and deep, and the x descent may also become prominent (w waveform). In pericardial tamponade, the x descent is prominent, but the y descent is diminished or absent.

EXAMINATION OF THE JUGULAR VENOUS PULSATIONS

EXAMINATION OF THE ARTERIAL PULSE

The examination of the neck veins allows for estimation of the right atrial pressure and for identification of the venous waveforms. The right internal jugular vein is used for this examination because it more accurately reflects right atrial pressure than the external jugular or left jugular vein. With the patient lying at a 45-degree angle (higher in patients with elevated venous pressure, lower in patients with low venous pressure) with the head turned to the left, the vertical distance from the sternal angle (angle of Louis) to the top of the venous pulsation can be determined. Because the right atrium lies about 5 cm vertically below the sternal angle, distention of the internal jugular vein 4cm above the sternal angle reflects a right atrial pressure of 9cm H2O. The right atrial pressure is normally 5 to 9 cm H2O and is increased with congestive heart failure, tricuspid insufficiency or stenosis, and restrictive or constrictive heart disease. With inspiration, negative intrathoracic pressure develops, venous blood drains into the thorax, and the normal venous pressure falls; the opposite is true with expiration. This pattern is reversed (Kussmaul sign) in the setting of right ventricular heart failure, constrictive pericarditis, or restrictive myocardial disease. With right ventricular heart failure, the elevated venous pressure results in passive congestion of the liver. Pressure applied over the liver for 1 to 3 minutes in this setting results in an increase in the jugular venous pressure (hepatojugular reflux). The normal waveforms of the venous pulsation consist of the a, c, and v waves and the x and y descents; these waveforms are shown in Figure 4-1A and reflect events in the right side of the heart. The a wave results from atrial contraction. Subsequent atrial relaxation results in a decrease in the right atrial pressure, which is seen as the x descent. This descent is interrupted by the c wave, generated by the bulging of the tricuspid valve cusps into the right atrium during ventricular systole. As the atrial pressure increases owing to venous return, the v wave is generated. This wave is normally smaller than the a wave and is followed by the y descent as the tricuspid valve opens and blood flows from the right atrium to the right ventricle during diastole. Abnormalities of the venous waveforms reflect underlying structural, functional, or electrical abnormalities of the heart (see Fig. 4-1B through G). The a wave increases in any condition in which greater resistance to right atrial emptying occurs (e.g., tricuspid stenosis, right ventricular hypertrophy or failure, pulmonary hypertension). Cannon a waves are seen when the atrium contracts against a closed tricuspid valve, as occurs with complete heart block, with junctional or ventricular rhythms, and occasionally with ventricular pacemakers. The a wave is absent in atrial fibrillation. In tricuspid regurgitation, the v wave is prominent and may

The arterial blood pressure can be measured with the use of a sphygmomanometer. The cuff is applied to the upper arm, rapidly inflated to 30 mm Hg above the anticipated systolic pressure, and then slowly deflated (= 3mmHg/sec) while listening for the sounds produced by blood entering the previously occluded brachial artery (Korotkoff sounds). The pressure at which the first sound is heard (usually a clear, tapping sound) represents the systolic pressure. Diastolic pressure occurs at the point at which the Korotkoff sounds disappear. Normally, the pressure in both arms is the same (about 120/70mmHg), and the systolic pressure in the legs is 10 to 20 mm Hg higher. Asymmetrical arm pressures can result from atherosclerotic disease of the aorta, aortic dissection, and stenosis of the innominate or subclavian arteries. Coarctation of the aorta and severe atherosclerotic disease of the aorta or the femoral or iliac arteries can result in a lower blood pressure in the legs than in the arms. Aortic insufficiency is frequently associated with a leg pressure more than 20 mm Hg higher than the arm pressure (Hill sign). Use of a cuff that is too small for a patient’s arm will result in erroneously high pressure measurements. Similarly, a cuff that is too large results in erroneously low measurements. The arterial examination should include assessments of the carotid, radial, brachial, femoral, popliteal, posterior tibial, and dorsalis pedis pulses, although the carotid artery pulse most accurately reflects the central aortic pulse. The rhythm, strength, contour, and symmetry of the pulses should be noted. The normal arterial pulse (Fig. 4-2A) rises rapidly to a peak in early systole, plateaus, and then falls. The descending pressure wave is interrupted by the dicrotic notch, related to aortic valve closure. This normal pattern is altered in a variety of cardiovascular disease states (see Fig. 4-2B through F). The amplitude of the pulse increases in aortic insufficiency, anemia, pregnancy, and thyrotoxicosis and decreases in conditions such as hypovolemia, tachycardia, left ventricular failure, and severe mitral stenosis. Aortic insufficiency results in a bounding pulse (Corrigan pulse or water-hammer pulse), owing to an increased pulse pressure (the difference between systolic and diastolic pressure), and is accompanied by a multitude of abnormalities in the peripheral pulses that reflect this increased pulse pressure. Aortic stenosis characteristically results in an attenuated carotid pulse with a delayed upstroke (pulsus parvus et tardus) and may be associated with a palpable thrill over the aortic area (the carotid shudder). A bisferious pulse is commonly felt in the presence of pure aortic regurgitation and is characterized by two systolic peaks. The first peak is the percussion wave, resulting from the rapid ejection of a large volume of blood early in systole; the second peak is the tidal

Canadian Cardiovascular Society has provided a similar classification of functional status specifically in patients with angina pectoris. These tools are useful in that they allow a patient’s symptoms to be classified and then compared with their symptoms at a different point in time.

Chapter 4—Evaluation of the Patient with Cardiovascular Disease

QRS T

ECG S1

phono

A a

a wave caused by atrial contraction, v wave during ventricular systole

v c

Atrial fibrillation, no a wave present

a v

Enhanced a wave

C a

c-v Dominant c-v wave

Exaggerated x and y descents in constrictive pericarditis

cannon a

Exaggerated x descent and loss of y descent in tamponade

cannon a

a JVP

G P

ECG P

Figure 4-1 Normal and abnormal jugular venous pulse tracings. A, Normal jugular pulse tracing with simultaneous electrocardiogram (ECG) and phonocardiogram. B, Loss of the a wave in atrial fibrillation. C, Large a wave in tricuspid stenosis. D, Large c-v wave in tricuspid regurgitation. E, Prominent x and y descents in constrictive pericarditis. F, Prominent x descent and diminutive y descent in pericardial tamponade. G, Jugular venous pulse (JVP) tracing and simultaneous ECG during complete heart block demonstrating cannon a waves occurring when the atrium contracts against a closed tricuspid valve during ventricular systole.

wave, a reflected wave from the periphery. This bifid pulse may also be noted in hypertrophic cardiomyopathy in which the initial rapid upstroke of the pulse is cut short by the development of a left ventricular outflow tract obstruction, resulting in a fall in the pulse. The reflected wave again produces the second impulse. In severe left ventricular dysfunc-

tion, the intensity of the pulse may alternate from beat to beat (pulsus alternans), and in atrial fibrillation, the pulse intensity is variable. With inspiration, negative intrathoracic pressure is transmitted to the aorta, and the systolic pressure normally decreases by up to 10mmHg. Pulsus paradoxus is an exaggeration of this normal inspiratory fall in systolic

Section III—Cardiovascular Disease

120 mm Hg

80 mm Hg 150 mm Hg

B 30 mm Hg Wide pulse pressure 100 mm Hg

80 mm Hg Delayed peak, narrow pulse pressure variable

variable

Bi-phase peak

90 mm Hg Alternating higher and lower pressure 60 mm Hg

90 mm Hg 70 mm Hg

60 mm Hg Expiration

Inspiration

Expiration

Figure 4-2 Normal and abnormal carotid arterial pulse contours. A, Normal arterial pulse with simultaneous electrocardiogram (ECG). The dicrotic wave (D) occurs just after aortic valve closure. B, Wide pulse pressure in aortic insufficiency. C, Pulsus parvus et tardus (small amplitude with a slow upstroke) associated with aortic stenosis. D, Bisferious pulse with two systolic peaks, typical of hypertrophic obstructive cardiomyopathy or aortic insufficiency, especially if concomitant aortic stenosis is present. E, Pulsus alternans, characteristic of severe left ventricular failure. F, Paradoxic pulse (systolic pressure decrease of >10mmHg with inspiration), most characteristic of cardiac tamponade.

pressure and is characteristically seen with pericardial tamponade, although it may also occur as a result of severe obstructive lung disease, constrictive pericarditis, hypovolemic shock, and pregnancy. Atherosclerotic disease of the peripheral vascular system frequently accompanies coronary atherosclerosis; therefore, the presence of peripheral vascular disease warrants a search for symptoms or signs of coronary artery disease and vice versa. When atherosclerosis occurs in a peripheral artery to the lower extremity and impairs blood flow distally, the patient may complain of intermittent cramping in the buttocks, thigh, calf, or foot (claudication). Severe peripheral vascular disease may result in digital ischemia or necrosis, without or with associated erectile dysfunction (Leriche syndrome). The peripheral pulses should be palpated and the abdominal aorta assessed for enlargement in all cardiac patients; a pulsatile, expansile, periumbilical mass suggests the presence of an abdominal aortic aneurysm. With significant stenosis of the peripheral vasculature, the distal

pulses may be diminished or absent, and the blood flow through the stenotic artery may be audible (a bruit). With normal aging, the elastic arteries lose their compliance, and this change in physical property may obscure abnormal findings.

EXAMINATION OF THE PRECORDIUM Inspection and palpation of the precordium may yield valuable clues as to the existence of cardiac disease. Chest wall abnormalities should be noted, such as pectus excavatum, which may be associated with Marfan syndrome or mitral valve prolapse; pectus carinatum, which may be associated with Marfan syndrome; and kyphoscoliosis, which is occasionally a cause of secondary pulmonary hypertension and right ventricular heart failure. The presence of visible pulsations in the aortic (second right intercostal space and suprasternal notch), pulmonic (third left intercostal space), right ventricular (left parasternal region), and left ventricu-

Chapter 4—Evaluation of the Patient with Cardiovascular Disease lar (fourth to fifth intercostal space and left mid-clavicular line) regions should be noted and will help direct the palpation of the heart. Retraction of the left parasternal area may be seen with severe left ventricular hypertrophy, and systolic retraction of the chest wall at the cardiac apex or left axilla (Broadbent sign) is characteristic of constrictive pericarditis. Precordial palpation is best performed with the patient supine or in the left lateral position, with the examiner standing to the patient’s right side. In this position, firm placement of the examiner’s right hand over the patient’s lower left chest wall places the fingertips over the region of the cardiac apex and the palm over the region of the right ventricle. The normal cardiac apical impulse is a brief, discrete impulse (about 1 cm) located in the fourth to fifth intercostal space in the left mid-clavicular line generated as the left ventricle strikes the chest wall during early systole. In a patient with a structurally normal heart, the apex is the point of maximal impulse (PMI) of the heart against the chest wall. Enlargement of the left ventricle results in lateral displacement of the apical impulse, whereas chronic obstructive pulmonary disease may result in inferior displacement of the PMI. Volume overload states, such as aortic insufficiency and mitral regurgitation, produce ventricular enlargement primarily from dilation and result in a hyperdynamic apical impulse; that is, the impulse is brisk and increased in amplitude. Pressure overload states, such as aortic stenosis and long-standing hypertension, produce ventricular enlargement primarily from hypertrophy. In this setting, the apical impulse is sustained, and atrial contraction is frequently detected (a palpable S4). Hypertrophic cardiomyopathy characteristically produces a double or triple apical impulse. Left ventricular aneurysms produce an apical impulse that is larger than normal and dyskinetic. The right ventricular impulse is not normally palpable. When an impulse is felt over the left parasternal region, it usually reflects right ventricular hypertrophy or dilation. Aortic aneurysms may be palpable (or visible) in the suprasternal notch or the second right intercostal space. Pulmonary hypertension may produce a palpable systolic impulse in the left third intercostal space and may also be associated with a palpable pulmonic component of the second heart sound (P2). Harsh murmurs originating from valvular or congenital heart disease may be associated with palpable vibratory sensations (thrills), as can occur with aortic stenosis, hypertrophic cardiomyopathy, and ventricular septal defects.

Auscultation TECHNIQUE Auscultation of the heart should ideally be performed in a quiet room with the patient in a comfortable position and the chest fully exposed. Certain heart sounds are better heard with either the bell or diaphragm of the stethoscope. Lowfrequency sounds are best heard with the bell applied to the chest wall with just enough pressure to form a seal. As more pressure is applied to the bell, low-frequency sounds are filtered out. High-frequency sounds are best heard with the diaphragm firmly applied to the chest wall. In a patient

with a normally situated heart, four major zones of cardiac auscultation are assessed. Aortic valvular events are best heard in the second right intercostal space. Pulmonary valvular events are best heard in the second left interspace. The fourth left interspace is ideal for auscultating tricuspid valvular events, and mitral valvular events are best heard at the cardiac apex or PMI. Because anatomic abnormalities, both congenital and acquired, can alter the location of the heart in the chest, the auscultatory areas may vary among patients. For instance, in patients with emphysema, the heart is shifted downward, and heart sounds may be best heard in the epigastrium. In dextrocardia, the heart lies in the right hemithorax, and the auscultatory regions are reversed. Additionally, auscultation in the axilla or supraclavicular areas or over the thoracic spine may be helpful in some settings, and having the patient lean forward, exhale, or perform various maneuvers may help accentuate particular heart sounds (Table 4-4).

NORMAL HEART SOUNDS The two major heart sounds heard during auscultation are termed S1 and S2. These heart sounds are high-pitched sounds originating from valve closure (Web Sounds, normal). S1 occurs at the onset of ventricular systole and corresponds to closure of the atrioventricular valves. It is usually perceived as a single sound, although occasionally its two components, M1 and T1, corresponding to closure of the mitral and tricuspid valves, respectively, can be heard. M1 occurs earlier, is the louder of the two components, and is best heard at the cardiac apex. T1 is somewhat softer and heard at the left lower sternal border. The second heart sound results from closure of the semilunar valves. The two components, A2 and P2, originating from aortic and pulmonic valve closure, respectively, can be easily distinguished. A2 is usually louder than P2 and is best heard at the right upper sternal border. P2 is loudest over the second left intercostal space. During expiration, the normal S2 is perceived as a single event. However, during inspiration, the augmented venous return to the right side of the heart and the increased capacitance of the pulmonary vascular bed result in a delay in pulmonic valve closure. In addition, the slightly decreased venous return to the left ventricle results in slightly earlier aortic valve closure. Thus, physiologic splitting of the second heart sound, with A2 preceding P2 during inspiration, is a normal respiratory event. Occasionally, additional heart sounds may be heard in normal individuals. A third heart sound (see later discussion) can be heard in normal children and young adults, in whom it is referred to as a physiologic S3; it is rarely heard after the age of 40 years in healthy individuals (Web Sounds, S3). A fourth heart sound (S4) is generated by forceful atrial contraction and is rarely audible in normal young indi viduals but is fairly common in older individuals (Web Sounds, S4). A murmur is an auditory vibration usually generated either by abnormally increased flow across a normal valve or by normal flow across an abnormal valve or structure. Innocent murmurs are always systolic murmurs, are usually soft and brief, and are by definition not associated with abnormalities of the cardiovascular system. They arise from

Section III—Cardiovascular Disease

Table 4-4 Effects of Physiologic Maneuvers on Auscultatory Events Maneuver

Major Physiologic Effects

Useful Auscultatory Changes

Respiration

↑ Venous return with inspiration

Valsalva (initial ↑ BP, phase I; followed by ↓ BP, phase II)

↓ BP, ↓ venous return, ↓ LV size (phase II)

Standing

↑ Venous return ↑ LV size

Squatting

↑ ↑ ↑ ↑ ↑

Venous return Systemic vascular resistance LV size Arterial pressure Cardiac output

↑ ↑ ↓ ↑ ↓ ↑ ↑ ↓

Ventricular filling Contractility Arterial pressure Cardiac output LV size Arterial pressure Cardiac output LV size

↑ Right heart murmurs and gallops with inspiration; splitting of S2 (see Fig. 4-3) ↑ HCM ↓ AS, MR MVP click earlier in systole; murmur prolongs ↑ HCM ↓ AS, MR MVP click earlier in systole; murmur prolongs ↑ AS, MR, AI ↓ HCM MVP click delayed; murmur shortens ↑ Gallops ↑ MR, AI, MS ↓ AS, HCM ↑ AS Little change in MR ↑ HCM, AS, MS ↓ AI, MR, Austin Flint murmur MVP click earlier in systole; murmur prolongs ↑ MR, AI ↓ AS, HCM MVP click delayed; murmur shortens

Isometric exercise (e.g., handgrip)

Post PVC or prolonged R-R interval Amyl nitrate

Phenylephrine

↑, Increased intensity; ↓, decreased intensity; AI, aortic insufficiency; AS, aortic stenosis; BP, blood pressure; HCM, hypertrophic cardiomyopathy; LV, left ventricle; MR, mitral regurgitation; MS, mitral stenosis; MVP, mitral valve prolapse; PVC, premature ventricular contraction; R-R, interval between the R waves on an electrocardiogram.

flow across the normal aortic or pulmonic outflow tracts and are present in a large proportion of children and young adults. Murmurs associated with high-flow states (e.g., pregnancy, anemia, fever, thyrotoxicosis, exercise) are not considered innocent, although they are not usually associated with structural heart disease. These are termed physiologic murmurs, owing to their association with altered physiologic states. Diastolic murmurs are never innocent or physiologic.

ABNORMAL HEART SOUNDS Abnormalities of S1 and S2 relate to abnormalities in their intensity (Table 4-5) or abnormalities in their respiratory splitting (Table 4-6). As noted, splitting of the S1 is normal but not frequently noted. This splitting becomes more apparent with right bundle branch block or with Ebstein anomaly of the tricuspid valve, owing to delay in closure of the tricuspid valve in these conditions (Web Sounds, Ebstein). The intensity of S1 is determined in part by the opening state of the atrioventricular valves at the onset of ventricular systole. If the valves are still widely open, as may occur with tachycardia or a short P-R interval, S1 will be accentuated. Conversely, in the presence of a long P-R interval, the mitral valve drifts toward a closed position before the onset of ventricular systole, and the subsequent S1 is soft. The intensity of S1 may vary in the presence of Mobitz type I heart block, atrioventricular dissociation, and atrial fibrillation when the relationship between atrial and ventricular systole varies. In mitral

stenosis with a pliable valve, the persistent pressure gradient at the end of diastole keeps the mitral valve leaflets relatively open and results in a loud S1 at the onset of systole. In severe mitral stenosis, when the mitral valve is heavily calcified and has decreased leaflet excursion, S1 becomes faint or absent (Figs. 4-3 and 4-4). S2 may be loud in systemic hypertension, owing to accentuated aortic valve closure (loud A2), or in pulmonary hypertension, owing to accentuated pulmonic valve closure (loud P2). When the aortic or pulmonary valves are stenotic, the force of valve closure is decreased, thus A1 and P2 become soft or inaudible. In this setting, S2 may appear to be single; in the setting of aortic stenosis, prolonged left ventricular ejection narrows the normal splitting of S2; and with severe aortic stenosis, S2 may become absent altogether as prolonged ejection and its accompanying murmur obscure P2. Wide splitting of the S2 with normal respiratory variation occurs when either pulmonic valve closure is delayed (e.g., right bundle branch block, pulmonic stenosis) or aortic valve closure occurs earlier owing to more rapid ejection of left ventricular volume (e.g., mitral regurgitation, ventricular septal defect). Fixed splitting of S2 without respiratory variation is characteristic of atrial septal defects and also occurs with right ventricular failure (Web Sounds, ASD). Paradoxic splitting of S2 is a reversal of the usual closure sequence of the aortic and pulmonic valves (i.e., P2 precedes A2). In this setting, a single S2 with inspiration and splitting of S2 with expiration can be heard. This circumstance occurs most commonly when delay occurs in closure of the aortic valve resulting from either delay in electrical conduction to

Chapter 4—Evaluation of the Patient with Cardiovascular Disease

Table 4-5 Abnormal Intensity of Heart Sounds S1

Loud

Short PR interval Mitral stenosis with pliable valve

Pulmonary hypertension Thin chest wall

Soft

Long PR interval Mitral regurgitation Poor left ventricular function Mitral stenosis with rigid valve Thick chest wall Atrial fibrillation Heart block

Systemic hypertension Aortic dilation Coarctation of the aorta Calcific aortic stenosis Aortic regurgitation

—

Varying

Valvular or subvalvular pulmonic stenosis

Table 4-6 Abnormal Splitting of S2 Single S2

Widely Split S2 with Normal Respiratory Variation

— Pulmonic stenosis

Right bundle branch block Left ventricular pacing

Systemic hypertension Coronary artery disease Any condition that can lead to paradoxical splitting of S2

Pulmonic stenosis Pulmonary embolism Idiopathic dilation of the pulmonary artery

Fixed Split S2

Paradoxically Split S2

Atrial septal defect Severe right ventricular dysfunction — — —

Left bundle branch block Right ventricular pacing Angina, myocardial infarction Aortic stenosis Hypertrophic cardiomyopathy Aortic regurgitation

Mitral regurgitation Ventricular septal defect

the left ventricle (e.g., left bundle branch block) or prolonged mechanical contraction of the left ventricle (e.g., aortic stenosis, hypertrophic cardiomyopathy). The third heart sound, S3 (also called the ventricular diastolic gallop), is a low-pitched sound occurring shortly after A2 in mid-diastole and heard best at the cardiac apex with the patient in the left lateral position. A pathologic S3 is distinguished from a physiologic S3 by age or the presence of underlying cardiac disease. It is frequently heard with ventricular systolic dysfunction from any cause and likely results either from blood entering the ventricle during the rapid filling phase of diastole or from the impact of the ventricle against the chest wall. Maneuvers that increase venous return accentuate S3, and maneuvers that decrease venous return make the S3 softer. An S3 can also be heard in hyperdynamic states, where it likely results from rapid early diastolic filling. The left ventricular S3 is best noticed at the cardiac apex, whereas the right ventricular S3 is heard best at the left lower sternal border and increases in intensity with inspiration. The timing of the S3 is similar to the sound generated by atrial tumors (tumor plop) and constrictive pericarditis (pericardial knock) and can also be confused with the opening snap of a stenotic mitral valve. The fourth heart sound, S4 (also called the atrial diastolic gallop), is best heard at the cardiac apex with the bell of the stethoscope. It is a low-pitched sound originating from the active ejection of blood from the atrium into a noncompliant ventricle and is therefore not present in the setting of atrial fibrillation. S4 is commonly heard in patients

with left ventricular hypertrophy from any cause (e.g., hypertension, aortic stenosis, hypertrophic cardiomyopathy) or acute myocardial ischemia and in hyperkinetic states. Frequently, the S4 is also palpable at the cardiac apex. S3 and S4 are occasionally present in the same patient. In the presence of tachycardia or a prolonged PR interval, the S3 and S4 may merge to produce a summation gallop. The opening of normal cardiac valves is not audible. However, abnormal valves may produce opening sounds. In the presence of a bicuspid aortic valve or in aortic stenosis with pliable valve leaflets, an ejection sound is audible as the leaflets open to their maximal extent. A similar ejection sound may originate from a stenotic pulmonic valve, and in this case, the ejection sound decreases in intensity with inspiration. These ejection sounds are high pitched, occur early in systole, and are frequently followed by the typical ejection murmur of aortic or pulmonic stenosis. Ejection sounds are also heard with systemic or pulmonary hypertension, the exact mechanism of which is not clear. Ejection sounds heard in mid to late systole are referred to as systolic clicks and are most commonly associated with mitral valve prolapse. As the redundant mitral valve prolapses and reaches its maximal superior displacement, it produces a high-pitched click. Several clicks may be heard as various parts of the redundant valve prolapse (Web Sounds, MVP). Frequently, the click is followed by a mitral regurgitant murmur. Maneuvers that decrease venous return cause the clicks to occur earlier in systole and the murmur to become longer (see Table 4-4).

Section III—Cardiovascular Disease

ECG S1 S2 Loud S1

M1 T1

A2 P2

Loud S2 S1

S2 S3

S3 gallop

S1 S2

S4 S1

S4 gallop S3-4 Summation gallop

S2 Expiration

S2 A P

Physiologic splitting Inspiration

P Expiration

P Inspiration

S2 OS S3

First Second heart heart sound sound (S1) (S2) Figure 4-4 The relationship of extra heart sounds to the normal first (S1) and second (S2) heart sounds. S1 is composed of the mitral (M1) and tricuspid (T1) closing sounds, although it is frequently perceived as a single sound. S2 is composed of the aortic (A2) and pulmonic (P2) closing sounds, which are usually easily distinguished. A fourth heart sound (S4) is soft and low pitched and precedes S1. A pulmonic or aortic ejection sound (ES) occurs shortly after S1. The systolic click (C) of mitral valve prolapse may be heard in mid or late systole. The opening snap (OS) of mitral stenosis is high pitched and occurs shortly after S2. A tumor plop or pericardial knock occurs at the same time and can be confused with an OS or an S3, which is lower in pitch and occurs slightly later.

Table 4-7 Grading System for Intensity of Murmurs

Fixed splitting

Grade Grade Grade Grade Grade

1 2 3 4 5

Inspiration S1

Abnormally wide but physiologic splitting

Expiration S1

S1 ES

Grade 6

Barely audible murmur Murmur of medium intensity Loud murmur, no thrill Loud murmur with thrill Very loud murmur; stethoscope must be on the chest to hear it; may be heard posteriorly Murmur audible with stethoscope off the chest

Expiration S2

Paradoxical splitting

S1 Inspiration

First Second heart heart sound sound Figure 4-3 Abnormal heart sounds can be related to abnormal intensity, abnormal presence of a gallop rhythm, or abnormal splitting of S2 with respiration. ECG, electrocardiogram.

The opening of abnormal mitral or tricuspid valves can also be heard in the presence of rheumatic valvular stenosis, when the sound is referred to as an opening snap (Web Sounds, MS). The snap is heard only if the valve leaflets are pliable and is generated as the leaflets abruptly dome during early diastole. The interval between S2 and the opening snap is of diagnostic importance; as the stenosis worsens and the atrial pressure increases, the mitral valve

opens earlier in diastole, and the interval between the S2 and the opening snap shortens.

MURMURS As stated previously, murmurs are a series of auditory vibrations generated when either abnormal blood flow across a normal cardiac structure or normal flow across an abnormal cardiac structure results in turbulent flow. These sounds are longer than the individual heart sounds and can be described by their location, intensity, frequency (pitch), quality, duration, and timing in relation to systole or diastole. The intensity of a murmur is graded on a scale of 1 to 6 (Table 4-7). In general, murmurs of grade 4 or greater are associated with a palpable thrill. The loudness of a murmur does not necessarily correlate with the severity of the underlying abnormality. For instance, flow across a large atrial septal defect is essentially silent, whereas flow across a small ventricular

Chapter 4—Evaluation of the Patient with Cardiovascular Disease

Aorta

AVO

MVO LA LV

S1 E Systolic ejection murmur Holosystolic regurgitant murmur

S1 E

Diastolic rumbling murmur of mitral stenosis

Decrescendo diastolic murmur

Figure 4-5 Abnormal sounds and murmurs associated with valvular dysfunction displayed simultaneously with left atrial (LA), left ventricular (LV), and aortic pressure tracings. AVO, aortic valve opening; E, ejection click of the aortic valve; MVO, mitral valve opening; OS, opening snap of the mitral valve. The shaded areas represent pressure gradients across the aortic valve during systole or mitral valve during diastole, characteristic of aortic stenosis and mitral stenosis, respectively.

septal defect is frequently associated with a loud murmur (Web Sounds, VSD). Higher-frequency murmurs correlate with a higher velocity of flow at the site of turbulence. Important to note are the pattern or configuration of the murmur (e.g., crescendo, crescendo-decrescendo, decrescendo, plateau) (Fig. 4-5) and the quality of the murmur (e.g., harsh, blowing, rumbling) as well as the location of maximal intensity and the pattern of radiation of the murmur. Various physical maneuvers may help clarify the nature of a particular murmur (see Table 4-4). Murmurs can be divided into three categories—(1) sys tolic, (2) diastolic, and (3) continuous (Table 4-8)—and can result from abnormalities on the right or left side of the heart as well as the great vessels. Right-sided murmurs may become significantly louder after inspiration, owing to the resulting augmentation of venous return, whereas left-sided murmurs are relatively unaffected by respiration. Systolic murmurs can be further divided into ejection-type murmurs and regurgitant murmurs. Ejection murmurs reflect turbulent flow across the aortic or pulmonic valve (Web Sounds, AS and PS). They begin shortly after S1, increase in intensity as the velocity of flow increases, and subsequently decrease in intensity as the velocity falls (crescendo-decrescendo). Examples of ejection-type murmurs include innocent murmurs and the murmurs of aortic sclerosis, aortic stenosis, pulmonic stenosis, and hypertrophic cardiomyopathy. Innocent murmurs and aortic sclerotic murmurs are short in duration and do not radiate (Web Sounds, benign murmur). The duration of aortic or pulmonic stenotic murmurs varies depending on the severity of the stenosis (compare Web Sounds, AS—early and AS—late). With

more severe stenosis, the murmur becomes longer, and the time to peak intensity of the murmur lengthens (i.e., early-, mid-, and late-peaking murmurs). The murmur of aortic stenosis is usually harsh, radiates to the carotid arteries, and at times may radiate to the cardiac apex (Gallavardin phenomenon). The murmur of hypertrophic cardiomyopathy may be confused with aortic stenosis, but it does not radiate to the carotids, and it is the only ejection murmur that becomes louder with decreased venous return. Mitral regurgitation associated with mitral valve prolapse may also show this response, but it is not a typical ejection murmur. The classic regurgitant systolic murmurs of mitral (MR) and tricuspid regurgitation (TR) last throughout all of systole (holosystolic), are plateau in pattern, and terminate at S2 (Web Sounds, MR). With acute MR, the murmur may be limited to early systole and may be somewhat decrescendo in pattern. When MR is secondary to mitral valve prolapse, it starts in mid to late systole and is preceded by a mitral valve click. Ventricular septal defects may also result in holosystolic murmurs, although a small muscular ventricular septal defect may have a murmur limited to early systole. Early-diastolic murmurs result from aortic or pulmonic insufficiency and are decrescendo in pattern. The duration of the murmur reflects chronicity: a short murmur is heard in acute aortic insufficiency or mild insufficiency, whereas chronic aortic insufficiency may produce a murmur throughout diastole. A Graham Steell murmur denotes a pulmonic insufficiency murmur in the setting of pulmonary hypertension. Mid-diastolic murmurs classically result from mitral or tricuspid stenosis, are low pitched, and are referred to as

Section III—Cardiovascular Disease

Table 4-8 Classification of Heart Murmurs Timing

Class

Description

Characteristic Lesions

Systolic

Ejection

Begins in early systole; may extend to mid or late systole Crescendo-decrescendo pattern Often harsh in quality Begins after S1 and ends before S2

Holosystolic

Extends throughout systole* Relatively uniform in intensity

Late

Variable onset and duration, often preceded by a nonejection click Begins with A2 or P2 Decrescendo pattern with variable duration Often high pitched, blowing Begins after S2, often after an opening snap Low-pitched rumble heard best with bell of stethoscope

Valvular, supravalvular, and subvalvular aortic stenoses Hypertrophic cardiomyopathy Pulmonic stenosis Aortic or pulmonary artery dilation Malformed but nonobstructive aortic valve ↑ Transvalvular flow (e.g., aortic regurgitation, hyperkinetic states, atrial septal defect, physiologic flow murmur) Mitral regurgitation Tricuspid regurgitation Ventricular septal defect Mitral valve prolapse

Diastolic

Early

Mid

Late Continuous

—

Louder with exercise and left lateral position Loudest in early diastole Presystolic accentuation of mid-diastolic murmur Systolic and diastolic components “machinery murmurs”

Aortic regurgitation Pulmonic regurgitation

Mitral stenosis Tricuspid stenosis ↑ Flow across atrioventricular valves (e.g., mitral regurgitation, tricuspid regurgitation, atrial septal defect) Mitral stenosis Tricuspid stenosis Patent ductus arteriosus Coronary atrioventricular fistula Ruptured sinus of Valsalva aneurysm into right atrium or ventricle Mammary souffle Venous hum

*Encompasses both the first and second heart sounds.

diastolic rumbles. Similar murmurs may be heard with obstructing atrial myxomas or in the presence of augmented diastolic flow across an unobstructed mitral or tricuspid valve, as occurs with an atrial or ventricular septal defect or with significant MR or TR. Severe, chronic aortic insufficiency may also produce a diastolic rumble, owing to premature closure of the mitral valve (Austin Flint murmur). Late-diastolic murmurs reflect presystolic accentuation of the mid-diastolic murmurs, owing to augmented mitral or tricuspid flow after atrial contraction. Continuous murmurs are murmurs that last throughout all of systole and continue into at least early diastole. These murmurs are referred to as machinery murmurs and are generated by continuous flow from a vessel or chamber with high pressure into a vessel or chamber with low pressure. A patent ductus arteriosus produces the classic continuous murmur (Web Sounds, PDA).

sternal border with the patient leaning forward and holding his or her breath at end expiration. The classic rub has three components corresponding to atrial systole, ventricular systole, and ventricular diastole, although frequently only one or two of the components are audible (Web Sounds, pericardial rubs). Localized irritation of the surrounding pleura may result in an associated pleural friction rub (pleuropericardial rub), which varies with respiration. Continuous venous murmurs, or venous hums, are almost universally present in children. They are also frequent in adults, especially during pregnancy or in the setting of thyrotoxicosis or anemia. These murmurs are best heard at the base of the neck with the patient’s head turned to the opposite direction and can be eliminated by gentle pressure over the vein.

OTHER CARDIAC SOUNDS

Prosthetic valves produce characteristic auscultatory findings. Porcine or bovine bioprosthetic valves produce heart sounds that are similar to native valve sounds; however, because these valves are smaller than the native valves that

Pericardial rubs occur in the setting of pericarditis. These rubs produce coarse, scratching sounds heard best at the left

PROSTHETIC HEART SOUNDS

Chapter 4—Evaluation of the Patient with Cardiovascular Disease they replace, they almost always have an associated murmur (systolic ejection murmur when placed in the aortic position and diastolic rumble when placed in the mitral position). Mechanical valves result in crisp, high-pitched sounds related to valvular opening and closure. With ball-in-cage valves (e.g., Starr-Edwards valves, the opening sound is louder than the closure sound. With all other mechanical

valves (e.g., Björk-Shiley valves, St. Jude valves), the closure sound is louder. These valves also produce an ejection-type murmur. Listening for all of the expected prosthetic sounds in patients with prosthetic valves is important because dysfunction of these valves may first be suggested by a change in the intensity or quality of the heart sounds or the development of a new or changing murmur.

Prospectus for the Future Thanks to advances in chip technology, the essential art of cardiac auscultation is making a resurgence with the use of the computerized heart sound phonocardiography or acoustic cardiography. Students and experienced practitioners alike will be able to use an algorithm for predicting left ventricular dysfunction based on the characteristics of the S3 and S4 heart sounds

References Pickering TG, Hall JE, Appel LJ, et al: Recommendations for blood pressure measurement in humans and experimental animals. Part 1: Blood pressure measurement in humans: A statement for professionals from the Subcommittee of Professional and Public Education of the American Heart Association Council on High Blood Pressure Research. Circulation 111:697-716, 2005.

and biomarkers of disease compensation and progression. Personal digital assistants, smartphones, and other technologies will make inroads for more accurate diagnosis during initial screening evaluation and bedside management of patients with cardiovascular disease.

Goldman L, Ausiello D: Cecil Textbook of Medicine, 22nd ed. Part VIII: Cardiovascular Disease. Philadelphia, WB Saunders, 2004.

III

Chapter

Diagnostic Tests and Procedures in the Patient with Cardiovascular Disease Sheldon E. Litwin

Chest Radiography The chest radiograph is an integral part of the cardiac evaluation and gives valuable information regarding structure and function of the heart, lungs, and great vessels. A routine examination includes posteroanterior and lateral projections (Fig. 5-1). In the posteroanterior view, cardiac enlargement may be present when the transverse diameter of the cardiac silhouette is greater than one half the transverse diameter of the thorax. The heart may appear falsely enlarged when it is displaced horizontally, such as with poor inflation of the lungs, and if the film is an anteroposterior projection, which magnifies the heart shadow. Left atrial enlargement is suggested when the left-sided heart border is straightened or bulges toward the left. In addition, the main bronchi may be widely splayed, and a circular opacity or double density within the cardiac silhouette may be seen. Right atrial enlargement may be present when the right-sided heart border bulges toward the right. Left ventricular enlargement results in downward and lateral displacement of the apex. A rounding of the displaced apex suggests ventricular hypertrophy. Right ventricular enlargement is best assessed in the lateral view and may be present when the right ventricular border occupies more than one third of the retrosternal space between the diaphragm and thoracic apex. The aortic arch and thoracic aorta may become dilated and tortuous in patients with severe atherosclerosis, longstanding hypertension, and aortic dissection. Dilation of the proximal pulmonary arteries may occur when pulmonary pressures are elevated and pulmonary vascular resistance is increased. Disease states associated with increased pulmonary artery flow and normal vascular resistance, such as 46

atrial or ventricular septal defects, may result in dilation of the proximal and distal pulmonary arteries. Pulmonary venous congestion secondary to elevated left ventricular heart pressures results in redistribution of blood flow in the lungs and prominence of the apical vessels. Transudation of fluid into the interstitial space may result in fluid in the fissures and along the horizontal periphery of the lower lung fields (Kerley B lines). As venous pressures further increase, fluid collects within the alveolar space, which early on collects preferentially in the inner two thirds of the lung fields, resulting in a characteristic butterfly appearance. Fluoroscopy or plain films may identify abnormal calcification involving the pericardium, coronary arteries, aorta, and valves. In addition, fluoroscopy can be instrumental in evaluating the function of mechanical prosthetic valves. Specific radiographic signs of congenital and valvular diseases are discussed in their respective sections.

Electrocardiography The electrocardiogram (ECG) represents the electrical activity of the heart recorded by skin electrodes. This wave of electrical activity is represented as a sequence of deflections on the ECG (Fig. 5-2). The horizontal scale represents time such that, at a standard paper speed of 25mm/second, each small box (1mm) represents 0.04 second, and each large box (5mm) represents 0.20 second. The vertical scale represents amplitude (10mm = 1mV). The heart rate can be estimated by dividing the number of large boxes between complexes (R-R interval) into 300. In the normal heart, the electrical impulse originates in the sinoatrial (SA) node and is conducted through the atria.

Chapter 5—Diagnostic Tests and Procedures in the Patient with Cardiovascular Disease

Figure 5-1 Schematic illustration of the parts of the heart, whose outlines can be identified on a routine chest radiograph. A, Posteroanterior chest radiograph. B, Lateral chest radiograph. Ao, aorta; LA, left atrium; LV, left ventricle; PA, pulmonary artery; RA, right atrium; RV, right ventricle.

0.2 sec 0.04 sec

1 mv

T P

PR interval

ST Q S segment QT interval

Figure 5-2 Normal electrocardiographic complex with labeling of waves and intervals.

Given that depolarization of the SA node is too weak to be detected on the surface ECG, the first, low-amplitude deflection on the surface ECG reflects atrial activation and is termed the P wave. The interval between the onset of the P wave and the next rapid deflection (QRS complex) is known as the PR interval and primarily represents the time taken for the impulse to travel through the atrioventricular (AV) node. The normal PR segment ranges from 0.12 to 0.20 second. A PR interval greater than 0.20 second defines AV nodal block. After the wave of depolarization has moved through the AV node, the ventricular myocardium is depolarized in a sequence of four phases. First, the interventricular septum depolarizes from left to right. This phase is followed by

depolarization of the right ventricle and inferior wall of the left ventricle, then the apex and central portions of the left ventricle, and, finally, the base and the posterior wall of the left ventricle. Ventricular depolarization results in a highamplitude complex on the surface ECG known as the QRS complex. The first downward deflection of this complex is the Q wave, the first upward deflection is the R wave, and the subsequent downward deflection is the S wave. In some individuals, a second upward deflection may be present after the S wave and is termed R prime (R′). Normal duration of the QRS complex is less than 0.10 second. Complexes greater than 0.12 second are usually secondary to some form of interventricular conduction delay. The isoelectric segment after the QRS complex is the ST segment and represents a brief period during which relatively little electrical activity occurs in the heart. The junction between the end of the QRS complex and the beginning of the ST segment is the J point. The upward deflection after the ST segment is the T wave and represents ventricular repolarization. The QT interval, which reflects the duration and transmural gradient of ventricular depolarization and repolarization, is measured from the onset of the QRS complex to the end of the T wave. The QT interval varies with heart rate, but for rates between 60 and 100 beats/ minute, the normal QT interval ranges from 0.35 to 0.44 second. For heart rates outside this range, the QT interval can be corrected by the following formula: QTc = QT (sec ) R-R interval1 2 (sec ) In some individuals, a U wave (of varying amplitude) may be noted after the T wave, the cause of which is unknown. The standard ECG consists of 12 leads: six limb leads (I, II, III, aVR, aVL, and aVF) and six chest or precordial leads (V1 to V6) (Fig. 5-3). The electrical activity recorded in each

Section III—Cardiovascular Disease

aVR

aVL

III

aVF

Figure 5-3 Normal 12-lead electrocardiogram.

Left-axis −90° −120°

dev ia

t io n(

−60°

−3 0°

−90 °)

) 0° −9 −30°

deviation (+90° to

−150° aVL

aVR

− +180°

0°

+30°

tio n

aVF

de via

III

+150°

+60°

+120° +90°

(−3 0° to +

axis

90°)

ht Rig

lead represents the direction and magnitude (vector) of the electrical force as seen from that particular lead position. Electrical activity directed toward a particular lead is represented as an upward deflection, and an electrical impulse directed away from a particular lead is represented as a downward deflection. Although the overall direction of electrical activity can be determined for any of the waveforms previously described, the mean QRS axis is the most clinically useful and is determined by examining the six limb leads. Figure 5-4 illustrates Einthoven triangle and the polarity of each of the six limb leads of the standard ECG. Skin electrodes are attached to both arms and legs, with the right leg serving as the ground. Leads I, II, and III are bipolar leads and represent electrical activity between two leads: lead I represents electrical activity between the right and left arms (left arm positive), lead II between the right arm and left leg (left leg positive), and lead III between the left arm and left leg (left leg positive). Leads aVR, aVL, and aVF are designated the augmented leads. With these leads, the QRS will be positive or have a predominant upward deflection when the electrical forces are directed toward the right arm for aVR, left arm for aVL, and left leg for aVF. These six leads form a hexaxial frontal plane of 30-degree arc intervals. The normal QRS axis ranges from −30 to +90 degrees. An axis more negative than −30 defines left axis deviation, and an axis greater than +90 defines right axis deviation. In general, a positive QRS complex in leads I and aVF suggests a normal QRS axis between 0 and 90 degrees.

is ax

Figure 5-4 Hexaxial reference figure for frontal plane axis determination, indicating values for abnormal left and right QRS axis deviations.

Chapter 5—Diagnostic Tests and Procedures in the Patient with Cardiovascular Disease

aVR

aVL

III

aVF

aVR

aVL

III

aVF

B Figure 5-5 A, Left ventricular hypertrophy as seen on an electrocardiographic recording. Characteristic findings include increased QRS voltage in precordial leads (deep S in lead V2 and tall R in lead V5) and downsloping ST depression and T-wave inversion in lateral precordial leads (strain pattern) and leftward axis. B, Right ventricular hypertrophy with tall R wave in right precordial leads, downsloping ST depression in precordial leads (RV strain), right axis deviation, and evidence of right atrial enlargement.

The six standard precordial leads (V1 to V6) are attached to the anterior chest wall (Fig. 5-5). Lead placement should be as follows: V1—fourth intercostal space, right sternal border; V2—fourth intercostal space, left sternal border; V3—midway between V2 and V4; V4—fifth intercostal space, left mid-clavicular line; V5—level with V4, left anterior axillary line; V6—level with V4, left mid-axillary line. The chest leads should be placed under the breast. Electrical activity directed toward these leads results in a positive deflection on the ECG tracing. Leads V1 and V2 are closest to the right ventricle and interventricular septum, and leads V5 and V6 are closest to the anterior and anterolateral walls of the left ventricle. Normally, a small R wave occurs in lead V1 reflecting septal depolarization and a deep S wave reflecting predominantly left ventricular activation. From V1 to V6, the R wave becomes larger (and the S

wave smaller) because the predominant forces directed at these leads originate from the left ventricle. The transition from a predominant S wave to a predominant R wave usually occurs between leads V3 and V4. Right-sided chest leads are used to look for evidence of right ventricular infarction. ST-segment elevation in V4R has the best sensitivity and specificity for making this diagnosis. For rightsided leads, standard V1 and V2 are switched, and V3R to V6R are placed mirror image to the standard left-sided chest leads. Some groups have advocated the use of posterior leads to increase the sensitivity for diagnosing lateral and posterior wall infarction or ischemia (areas that are often deemed to be electrically silent on traditional 12-lead ECGs). To do this, six additional leads are placed in the fifth intercostal space continuing posteriorly from the position of V6.

Section III—Cardiovascular Disease

Abnormal Electrocardiographic Patterns CHAMBER ABNORMALITIES AND VENTRICULAR HYPERTROPHY The P wave is normally upright in leads I, II, and F; inverted in aVR; and biphasic in V1. Left atrial abnormality (defined as enlargement, hypertrophy, or increased wall stress) is characterized by a wide P wave in lead II (0.12 second) and a deeply inverted terminal component in lead V1 (1 mm). Right atrial abnormality is present when the P waves in the limb leads are peaked and 2.5mm or more in height. Left ventricular hypertrophy may result in increased QRS voltage, slight widening of the QRS complex, late intrinsicoid deflection, left axis deviation, and abnormalities of the ST-T segments (see Fig. 5-5A). Multiple criteria with variable sensitivity and specificity for detecting left ventricular hypertrophy are available. The most frequently used criteria are given in Table 5-1. Right ventricular hypertrophy is characterized by tall R waves in leads V1 through V3; deep S waves in leads I, aVL, V5, and V6; and right axis deviation (see Fig. 5-5B). In patients with chronically elevated pulmonary pressures, such as with chronic lung disease, a combination of ECG abnormalities reflecting a right-sided pathologic condition may be present and include right atrial abnormality, right ventricular hypertrophy, and right axis deviation. In patients with acute pulmonary embolus, ECG changes may suggest right

Table 5-1 Electrocardiographic Manifestations of Atrial Abnormalities and Ventricular Hypertrophy Left Atrial Abnormality P-wave duration ≥ 0.12 second Notched, slurred P wave in leads I and II Biphasic P wave in lead V1 with a wide, deep, negative terminal component Right Atrial Abnormality P-wave duration ≤ 0.11 second Tall, peaked P waves of ≥ 2.5mm in leads II, III, and aVF Left Ventricular Hypertrophy Voltage criteria R wave in lead aVL ≥ 12mm R wave in lead I ≥ 15mm S wave in lead V1 or V2 + R wave in lead V5 or V6 ≥ 35mm Depressed ST segments with inverted T waves in the lateral leads Left axis deviation QRS duration ≥ 0.09 second Left atrial enlargement Right Ventricular Hypertrophy Tall R waves over right precordium (R : S ratio in lead V1 > 1.0) Right axis deviation Depressed ST segments with inverted T waves in leads V1 to V3 Normal QRS duration (if no right bundle branch block) Right atrial enlargement

ventricular strain and include right axis deviation; incomplete or complete right bundle branch block; S waves in leads I, II, and III; and T-wave inversions in leads V1 through V3.

INTERVENTRICULAR CONDUCTION DELAYS The ventricular conduction system consists of two main branches, the right and left bundles. The left bundle further divides into the anterior and posterior fascicles. Conduction block can occur in either of the major branches or in the fascicles (Table 5-2). Fascicular block results in a change in the sequence of ventricular activation but does not prolong overall conduction time (QRS duration remains < 0.10 second). Left anterior fascicular block is a relatively common ECG abnormality and is sometimes associated with right bundle branch block. This conduction abnormality is present when extreme left axis deviation occurs (more negative than −45 degrees); when the R wave is greater than the Q wave in leads I and aVL; and when the S wave is greater than the R wave in leads II, III, and aVF. Left posterior fascicular block is uncommon but is associated with right axis deviation (>90 degrees); small Q waves in leads II, III, and aVF; and small R waves in leads I and aVL. The ECG findings associated with fascicular blocks can be confused with myocardial infarction (MI). For example, with left anterior fascicular block, the prominent QS deflection in leads V1 and V2 can mimic an anteroseptal MI, and the rS deflection in leads II, III, and aVF can be confused with an inferior MI. Similarly, the rS deflection in leads I and aVL in left posterior fascicular block may be confused with a high lateral infarct. The presence of abnormal ST- and T-wave segments and pathologic Q waves (see

Table 5-2 Electrocardiographic Manifestations of Fascicular and Bundle Branch Blocks Left Anterior Fascicular Block QRS duration ≤ 0.1 second Left axis deviation (more negative than −45 degrees) rS pattern in leads II, III, and aVF qR pattern in leads I and aVL Right Posterior Fascicular Block QRS duration ≤ 0.1 second Right axis deviation (+90 degrees or greater) qR pattern in leads II, III, and aVF rS pattern in leads I and aVL Exclusion of other causes of right axis deviation (chronic obstructive pulmonary disease, right ventricular hypertrophy) Left Bundle Branch Block QRS duration ≥ 0.12 second Broad, slurred, or notched R waves in lateral leads (I, aVL, V5, and V6) QS or rS pattern in anterior precordium leads (V1 and V2) ST-T-wave vectors opposite to terminal QRS vectors Right Bundle Branch Block QRS duration ≥ 0.12 second Large R′ wave in lead V1 (rsR′) Deep terminal S wave in lead V6 Normal septal Q waves Inverted T waves in leads V1 and V2

Chapter 5—Diagnostic Tests and Procedures in the Patient with Cardiovascular Disease “Myocardial Ischemia and Infarction” later) are helpful findings to differentiate MI from a fascicular block. Bundle branch blocks are associated with QRS duration longer than 120 milliseconds. In left bundle branch block, depolarization proceeds down the right bundle, across the interventricular septum from right to left, and then to the left ventricle. Characteristic ECG findings include a wide QRS complex; a broad R wave in leads I, aVL, V5, and V6; a deep QS wave in leads V1 and V2; and ST depression and T-wave inversion opposite the QRS deflection (Fig. 5-6A). Given the abnormal sequence of ventricular activation with left bundle branch block, many ECG abnormalities, such as Q-wave MI and left ventricular hypertrophy, are difficult to evaluate. In some cases, acute MI is still apparent even with LBBB. Left bundle branch block almost always indicates the presence of underlying myocardial disease (most commonly fibrosis due to ischemic injury or hypertrophy). With right bundle branch block, the interventricular septum depolarizes normally from left to right, and therefore the initial QRS deflection remains unchanged. As a result, ECG abnormalities such as Q-wave MI can still be interpreted. After septal activation, the left ventricle depolarizes, A

followed by the right ventricle. The ECG is characterized by a wide QRS complex; a large R′ wave in lead V1 (R-S-R′); and deep S waves in leads I, aVL, and V6, representing delayed right ventricular activation (see Fig. 5-6B). Although right bundle branch block may be associated with underlying cardiac disease, it may also appear as a normal variant or be seen intermittently when heart rate is elevated. In the latter case, it is often referred to as rate-related bundle branch block.

MYOCARDIAL ISCHEMIA AND INFARCTION Myocardial ischemia and MI may be associated with abnormalities of the ST segment, T-wave, and QRS complex. Myocardial ischemia primarily affects repolarization of the myocardium and is often associated with horizontal or down-sloping ST-segment depression and T-wave inversion. These changes may be transient, such as during an anginal episode or an exercise stress test, or may be longlasting in the setting of unstable angina or MI. T-wave inversion without ST-segment depression is a nonspecific finding and must be correlated with the clinical setting. B

Left bundle branch block

Right bundle branch block

III

aVR

aVL

aVF

aVR

aVL

aVF

Diagnostic criteria for LBBB QRS duration > 0.125 Broad R wave in I1aVL,V5–V6 Deep as in V1–V2 T-wave inversion in lateral leads

Diagnostic criteria for RBBB QRS duration > 0.125 R > S in V1 RSR in V1 Deep wide S wave I and V6

Figure 5-6 A, Left bundle branch block (LBBB). B, Right bundle branch block (RBBB). Criteria for bundle branch block are summarized in Table 5-2.

Section III—Cardiovascular Disease I

III

aVR

aVL

aVF

Control

2 hours later

24 hours later

48 hours later

8 days later

6 months later

aVR

aVL

III

aVF

III

Chapter 5—Diagnostic Tests and Procedures in the Patient with Cardiovascular Disease

Figure 5-7 A, Evolutionary changes in a posteroinferior myocardial infarction. Control tracing is normal. The tracing recorded 2 hours after onset of chest pain demonstrated development of early Q waves, marked ST-segment elevation, and hyperacute T waves in leads II, III, and aVF. In addition, a larger R wave, ST-segment depression, and negative T waves have developed in leads V1 and V2. These are early changes indicating acute posteroinferior myocardial infarction. The 24-hour tracing demonstrates evolutionary changes. In leads II, III, and aVF, the Q wave is larger, the ST segments have almost returned to baseline, and the T wave has begun to invert. In leads V1 to V2, the duration of the R wave now exceeds 0.04 seconds, the ST segment is depressed, and the T wave is upright. (In this example, ECG changes of true posterior involvement extend past lead V2; ordinarily, only leads V1 and V2 may be involved.) Only minor further changes occur through the 8-day tracing. Finally, 6 months later, the ECG illustrates large Q waves, isoelectric ST segments, and inverted T waves in leads II, III, and aVF and large R waves, isoelectric ST segment, and upright T waves in leads V1 and V2, indicative of an old posteroinferior myocardial infarction. B, Example of an ECG from a patient with an underlying LBBB who experienced an acute anterior myocardial infarction. Characteristic ST segment elevation and hyperacute T-waves are seen in leads V1-V6 and leads I and AVL despite the presence of the LBBB. Note that this is not always the case, as a patient with typical symptoms and a LBBB as well as no definite ischemic ST segment elevations should still be treated as if the individual is having an MI or acute coronary syndrome.

Localized ST-segment elevation suggests more extensive myocardial injury and is often associated with acute MI (see Fig. 5-7). Vasospastic or Prinzmetal angina may be associated with reversible ST-segment elevation without MI. STsegment elevation may occur in other settings not related to acute ischemia or infarction. Persistent, localized STsegment elevation in the same leads as pathologic Q waves is consistent with a ventricular aneurysm. Acute pericarditis is associated with diffuse ST-segment elevation and PR depression. Diffuse J-point elevation in association with upward-coving ST segments is a normal variant common among young men and is often referred to as early repolarization. The presence of a Q wave is one of the diagnostic criteria used to verify MI. Infarcted myocardium is unable to conduct electrical activity, and therefore electrical forces will be directed away from the surface electrode overlying the infarcted region, resulting in a Q wave on the surface ECG. Knowing which region of the myocardium each lead represents enables the examiner to localize the area of infarction (Table 5-3). A pathologic Q wave has a duration of greater than or equal to 0.04 second or a depth one fourth or more the height of the corresponding R wave. Not all MIs result in the formation of Q waves. In addition, small R waves can return many weeks to months after an MI. Abnormal Q waves, or pseudoinfarction, may also be associated with nonischemic cardiac disease, such as ventricular pre-excitation, cardiac amyloidosis, sarcoidosis, idiopathic or hypertrophic cardiomyopathy, myocarditis, and chronic lung disease.

ABNORMALITIES OF THE ST SEGMENT AND T WAVE A number of drugs and metabolic abnormalities may affect the ST segment and T wave (Fig. 5-8). Hypokalemia may result in prominent U waves in the precordial leads and prolongation of the QT interval. Hyperkalemia may result in tall, peaked T waves. Hypocalcemia typically lengthens the QT interval, whereas hypercalcemia shortens it. A commonly used cardiac medication, digoxin, often results in diffuse, scooped ST-segment depression. Minor or nonspecific ST-segment and T-wave abnormalities may be present in many patients and have no definable cause. In these instances, the physician must determine the significance of the abnormalities based on the clinical setting. Several excellent websites containing examples of normal and abnormal ECGs are available.

Long-Term Ambulatory Electrocardiographic Recording Ambulatory ECG (Holter monitoring) is a widely used, noninvasive method to evaluate cardiac arrhythmias and conduction disturbances over an extended period and to detect electrical abnormalities that may be brief or transient. With this approach, ECG data from two to three surface leads are stored on a tape recorder that the patient wears for a minimum of 24 to 48 hours. The recorders have both patient-activated event markers and time markers so that any abnormalities can be correlated with the patient’s symptoms or time of day. These data can then be printed in a standard, real-time ECG format for review. For patients with intermittent or rare symptoms, an event recorder, which can be worn for several weeks, may be helpful in identifying the arrhythmia. The simplest device is a small, hand-held monitor that is applied to the chest wall when symptoms occur. The ECG data are recorded and can be transmitted later by telephone to a monitoring center for analysis. A more sophisticated system uses a wrist recorder that allows continuous-loop storage of 4 to 5 minutes of ECG data from one lead. When the patient activates the system, ECG data preceding the event and for 1 to 2 minutes after the event are recorded and stored for further analysis. With both of these devices, the patient must be physically able to activate the recorder during the episode to store the ECG data. Implantable recording devices (subcutaneous) are sometimes used to diagnose infrequent events.

STRESS TESTING Stress testing is an important noninvasive tool for evaluating patients with known or suggested coronary artery disease (CAD). During exercise, the increased demand for oxygen by the working skeletal muscles is met by increases in heart rate and cardiac output. In patients with significant CAD, the increase in myocardial oxygen demand cannot be met by an increase in coronary blood flow. As a result, myocardial ischemia may occur, resulting in chest pain and characteristic ECG abnormalities. These changes, combined with the hemodynamic response to exercise, can give useful diagnostic and prognostic information in the patient with cardiac abnormalities. The most frequent indications for stress testing include establishing a diagnosis of CAD in patients

Section III—Cardiovascular Disease

Table 5-3 Electrocardiographic Localization of Myocardial Infarction Leads Depicting Primary Electrocardiographic Changes

Infarct Location Inferior Septal Anterior Anteroseptal Extensive anterior Lateral High lateral Posterior† Right ventricular‡

Likely Vessel* Involved

II, III, aVF V1, V2 V3, V4 V1 to V4 I, aVL, V1 to V6 I, aVL, V5 to V6 I, aVL Prominent R in V1 ST elevation in V1 and, more specifically, V4R in setting of inferior infarction

RCA LAD LAD LAD LAD CIRC CIRC RCA or CIRC RCA

*This is a generalization; variations occur. † Usually in association with inferior or lateral infarction. ‡ Usually in association with inferior infarction. CIRC, circumflex artery; LAD, left anterior descending coronary artery; RCA, right coronary artery.

Normal

Hyperkalemia

Hypokalemia

Hypercalcemia Hypocalcemia Hypothermia

Digitalis

Quinidine Procainamide Disopyramide Phenothiazines Tricyclic antidepressants CNS insult (e.g., intracerebral hemorrhage)

Mild to moderate (K = 5-7 mEq/L): Tall, symmetrically peaked T waves with a narrow base More severe (K = 8-11 mEq/L): QRS widens, PR segment prolongs, P wave disappears; ECG resembles a sine wave in severe cases ST depression T-wave flattening Large positive U wave, QT prolongation due to U wave Shortened QT interval due to a shortened ST segment Prolonged QT interval due to a prolonged ST segment; T-wave duration normal Osborne or J waves: J-point elevation with a characteristic elevation of the early ST segment. Slow rhythm, baseline artifact due to shivering often present. ST depression T-wave flattening or inversion Shortened QT interval, increased Uwave amplitude Prolonged QT interval, mainly due to prolonged T-wave duration with flattening or inversion QRS prolongation Increased U-wave amplitude Diffuse, wide, deeply inverted T waves with prolonged QT

Figure 5-8 Metabolic and drug influences on the electrocardiographic recording.

with chest pain, assessing prognosis and functional capacity in patients with chronic stable angina or after an MI, evaluating exercise-induced arrhythmias, and assessing for ischemia after a revascularization procedure. The most common form of stress testing uses continuous ECG monitoring while the patient walks on a treadmill.

With each advancing stage, the speed and incline of the belt increases, thus increasing the amount of work the patient performs. The commonly used Bruce protocol employs 3 minutes of exercise at each stage. The modified Bruce protocol incorporates two beginning stages with slower speeds and lesser inclines than are used in the standard Bruce pro-

Chapter 5—Diagnostic Tests and Procedures in the Patient with Cardiovascular Disease tocol. The modified Bruce or similar protocols are generally used for older, markedly overweight, and unstable or more debilitated patients. Exercise testing may also be performed using a bicycle or arm ergometer. The stress test is generally deemed adequate if the patient achieves 90% of his or her predicted maximal heart rate, which is equal to 220 minus the patient’s age. Indications for stopping the test include fatigue, severe hypertension (>220mmHg systolic), worsening angina during exercise, developing marked or widespread ischemic ECG changes, significant arrhythmias, or hypotension. The diagnostic accuracy of stress testing is improved with adjunctive echocardiography or radionuclide imaging. Contraindications to stress testing include unstable angina, acute MI, poorly controlled hypertension (blood pressure >220/110mmHg), severe aortic stenosis (valve area < 1.0cm2), and decompensated congestive heart failure. In the era of reperfusion therapy (thrombolytic and percutaneous interventions), for acute coronary syndromes or acute MI, little role exists for the predischarge submaximal stress test that was commonly used in the past. The diagnostic accuracy of the exercise test is dependent on the pre-test likelihood of CAD in a given patient, the sensitivity and specificity of the test results in that patient population, and the ECG criteria used to define a positive test. Clinical features that are most useful at predicting important angiographic coronary disease before exercise testing include advanced age, male sex, and the presence of typical (vs. atypical) anginal chest pain. The diagnostic accuracy and cost-effectiveness of exercise testing is best in patients with an intermediate risk for CAD (30% to 70%) and when ischemic ECG changes are accompanied by chest pain during exercise. Exercise testing is less cost-effective in diagnosing CAD in a patient with classic symptoms of angina because a positive test will not significantly increase the post-test probability of CAD, and a negative test would likely represent a false-negative result. Nonetheless, prognostic information and objective information about the efficacy of pharmacologic therapy may still be obtained. Similarly, exercise testing in young patients with atypical chest pain may not be diagnostically useful, given that an abnormal test result will likely represent a false-positive test and will not significantly increase the post-test probability of CAD. The normal physiologic response to exercise is an increase in heart rate and systolic and diastolic blood pressures. The ECG will maintain normal T-wave polarity, and the ST segment will remain unchanged or, if depressed, will have a rapid upstroke back to baseline. An ischemic ECG response to exercise is defined as (1) 1.5 mm of up-sloping STsegment depression measured 0.08 second past the J point, (2) at least 1mm of horizontal ST depression, or (3) 1mm of down-sloping ST-segment depression measured at the J point. Given the large amount of artifact on the ECG that may occur with exercise, these changes must be present in at least three consecutive depolarizations. Other findings suggestive of more extensive CAD include early onset of ST depression (6 minutes); marked, down-sloping ST depression (>2mm), especially if present in more than five leads; ST changes persisting into recovery for more than 5 minutes; and failure to increase systolic blood pressure to 120mm Hg or more or a sustained decrease of 10mm Hg or more below baseline.

The ECG is not diagnostically useful in the presence of left ventricular hypertrophy, left bundle branch block, WolffParkinson-White syndrome, or chronic digoxin therapy. In these instances, nuclear or echocardiographic imaging is needed to diagnose ischemia. In patients who are unable to exercise, pharmacologic stress testing with myocardial imaging has been shown to have sensitivity and specificity for detecting CAD equal to those of exercise stress imaging. Intravenous dipyridamole and adenosine and newer selective adenosine A2A receptor agonists are coronary vasodilators that result in increased blood flow in normal arteries without significantly changing flow in diseased vessels. The resulting heterogeneity in blood flow can be detected by nuclear imaging techniques and the regions of myocardium supplied by diseased vessels identified. Another commonly used technique to evaluate for ischemia is dobutamine-stress echocardiography. Dobutamine is an inotropic agent that increases myocardial oxygen demand by increasing heart rate and contractility. The echocardiogram is used to monitor for ischemia, which is defined as new or worsening wall motion abnormalities during the infusion. Demon strating improvement in wall thickening with low-dose dobutamine suggests that there is myocardial viability of abnormal segments (i.e., segments that are hypokinetic or akinetic at baseline).

ECHOCARDIOGRAPHY Echocardiography is a widely used, noninvasive technique in which sound waves are used to image cardiac structures and evaluate blood flow. A piezoelectric crystal housed in a transducer placed on the patient’s chest wall produces ultrasound waves. As the sound waves encounter structures with different acoustic properties, some of the ultrasound waves are reflected back to the transducer and recorded. Ultrasound waves emitted from a single, stationary crystal produce an image of a thin slice of the heart (M mode), which can then be followed through time. Steering the ultrasound beam across a 90-degree arc multiple times per second creates two-dimensional imaging (Fig. 5-9). Transthoracic echocardiography is safe, simple, fast, and relatively inexpensive. Hence it is the most commonly used test to assess cardiac size, structure, and function. The development of three-dimensional echocardiographic imaging techniques offers great promise for more accurate measurements of chamber volumes and mass as well as the assessment of geometrically complex anatomy and valvular lesions (Web Fig. 5-1 shows a three-dimensional image). Doppler echocardiography allows assessment of both direction and velocity of blood flow within the heart and great vessels. When ultrasound waves encounter moving red blood cells, the energy reflected back to the transducer is altered. The magnitude of this change (Doppler shift) is represented as velocity on the echocardiographic display and can be used to determine whether the blood flow is normal or abnormal (Fig. 5-10). In addition, the velocity of a particular jet of blood can be converted to pressure using the modified Bernoulli equation (ΔP ≅ 4v2). This process allows for the assessment of pressure gradients across valves or between chambers. Color Doppler imaging allows visualization of blood flow through the heart by assigning a color to the red blood cells based on their velocity and direction

Section III—Cardiovascular Disease

RV LV

IVS

Figure 5-9 Portions of standard two-dimensional echocardiograms (A, parasternal long-axis view; B, apical four-chamber view) showing the major cardiac structures. Ao, aorta; IVS, interventricular septum; LA, left atrium; LV, left ventricle; MV, mitral valve; PE, pericardial effusion; PW, posterior LV wall; RV, right ventricle. See Web Figure 5-3 for a moving image of a two-dimensional echocardiogram. (Image courtesy of Sheldon E. Litwin, MD, Division of Cardiology, University of Utah.)

Figure 5-10 Doppler tracing in a patient with aortic stenosis and regurgitation. The velocity of systolic flow is related to the severity of obstruction.

(Fig. 5-11; Web Fig. 5-2). By convention, blood moving away from the transducer is represented in shades of blue, and blood moving toward the transducer is represented in red. Color Doppler imaging is particularly useful in identifying valvular insufficiency and abnormal shunt flow between chambers. Recently, the use of Doppler techniques to record myocardial velocities or strain rates has provided new insight into myocardial function and hemodynamics. Two-dimensional echocardiography and Doppler echo cardiography are often used in conjunction with exercise or pharmacologic stress testing. Although variability occurs among studies, the sensitivity of stress echocardiography is apparently slightly lower, but the specificity is slightly

higher, compared with myocardial perfusion imaging with nuclear tracers. The overall cost-effectiveness of stress echocardiography is estimated to be significantly better than nuclear perfusion imaging because of the lower cost. The development of ultrasound contrast agents composed of microbubbles that are small enough to transit through the pulmonary circulation has greatly improved the ability to use ultrasound to image obese patients, patients with lung disease, and those with otherwise difficult acoustic windows (Fig. 5-12; Web Fig. 5-3 shows a dynamic contrast echocardiographic image). These agents are also being developed as molecular imaging agents by complexing the bubbles to

Chapter 5—Diagnostic Tests and Procedures in the Patient with Cardiovascular Disease compounds that can selectively bind to the target site of interest (i.e., clots, neovessels). Transesophageal echocardiography (TEE) allows twodimensional and Doppler imaging of the heart through the esophagus by having the patient swallow a gastroscope mounted with an ultrasound crystal within its tip. Given

the close proximity of the esophagus to the heart, highresolution images can be obtained, especially of the left atrium, mitral valve apparatus, and aorta. TEE is particularly useful in diagnosing aortic dissection, endocarditis, prosthetic valve dysfunction, and left atrial masses (Fig. 5-13; Web Fig. 5-4).

NUCLEAR CARDIOLOGY Radionuclide imaging of the heart allows quantification of left ventricular size and systolic function as well as myocardial perfusion. With radionuclide ventriculography, the patient’s red blood cells are labeled with a small amount of a radioactive tracer (usually technetium-99m). Left ventricular function can then be assessed by one of two methods. With the first-pass technique, radiation emitted by the tagged red blood cells as they initially flow though the heart is detected by a gamma camera positioned over the patient’s chest. With the gated equilibrium method, or multigated acquisition (MUGA) method, the tracer is allowed to achieve an equilibrium distribution throughout the blood pool before count acquisition begins. This second method improves the resolution of the ventriculogram. For both techniques, the gamma camera can be gated to the ECG, allowing for determination of the total emitted end-diastole counts (EDC) and end-systole counts (ESC). Left ventricular ejection fraction (LVEF) can then be calculated as follows:

Figure 5-11 Color Doppler recording demonstrating severe mitral regurgitation. The regurgitant jet seen in the left atrium (LA) is represented in blue because blood flow is directed away from the transducer. The yellow components are the mosaic pattern traditionally assigned to turbulent or high-velocity flow. The arrow points to the hemisphere of blood accelerating proximal to the regurgitant orifice (proximal isovelocity surface area [PISA]). The size of the PISA can be used to help grade the severity of regurgitation. LA, left atrium; LV, left ventricle. See Web Figure 5-2 for a dynamic echocardiographic image in a patient with mitral regurgitation. (Image courtesy of Sheldon E. Litwin, MD, Division of Cardiology, University of Utah.)

LVEF = ( EDC − ESC ) EDC If scintigraphic information is collected throughout the cardiac cycle, a computer-generated image of the heart can be displayed in a cinematic fashion, allowing for the assessment of wall motion. Myocardial perfusion imaging is usually performed in conjunction with exercise or pharmacologic (vasodilator) stress testing. Persantine, or more commonly adenosine, is

Figure 5-12 Echocardiogram enhanced with intravenous ultrasound contrast agent (A, apical four-chamber view; B, apical long-axis view). Highly echo-reflectant microbubbles make the left ventricular cavity appear white, whereas the myocardium appears dark. See Web Figure 5-3 for a dynamic image of echocardiographic contrast. (Image courtesy of Sheldon E. Litwin, MD, Division of Cardiology, University of Utah.)

Section III—Cardiovascular Disease

LA V

V MV

Figure 5-13 Transesophageal echocardiogram demonstrating the presence of a vegetation adherent to the ring of a bi-leaflet tilting-disk mitral valve prostheses (A, systole, leaflets closed with vegetation seen in left atrium; B, diastole, leaflets open, vegetation prolapsing into left ventricle). Transesophageal echocardiography is the diagnostic test of choice for assessing prosthetic mitral valves because the esophageal window allows unimpeded views of the atrial surface of the valve. LA, left atrium; LV, left ventricle; MV, prosthetic mitral valve disks; V, vegetation. See Web Figure 5-4 for a dynamic transesophageal echocardiographic image. (Image courtesy of Sheldon E. Litwin, MD, Division of Cardiology, University of Utah.)

used as the coronary vasodilator. Each agent can increase myocardial blood flow by fourfold to fivefold. Adenosine is more expensive, but has the advantage over Persantine of a very short half-life. Newer adenosine-like agents with reduced side-effect profiles are starting to be used clinically. Technetium-99m sestamibi is the most frequently used radionuclide and is usually injected just before completion of the stress test. Tomographic (single-photon emission computed tomography [SPECT]) images of the heart are obtained for qualitative and quantitative analyses at rest and after stress. In the normal heart, radioisotope is relatively equally distributed throughout the myocardium. In patients with ischemia, a localized area of decreased uptake will occur after exercise but partially or completely fill in at rest (redistribution). A persistent defect at peak exercise and rest (fixed defect) is consistent with MI or scarring. However, in some patients with apparently fixed defects, repeat rest imaging at 24 hours or after reinjection of a smaller quantity of isotope will demonstrate improved uptake, indicating the presence of viable, but severely ischemic, myocardium. The use of new approaches such as combined low-level exercise and vasodilators, prone imaging, attenuation correction, and computerized data analysis has improved the quality and reproducibility of the data from these studies. Myocardial perfusion imaging may also be combined with ECG-gated image acquisition to allow for simultaneous assessment of ventricular function and perfusion. Not only can LVEF be quantitated with this technique, but also regional wall motion can be assessed to help rule out artifactual perfusion defects (Web Fig. 5-5). Positron-emission tomography (PET) is a noninvasive method of detecting myocardial viability by the use of both perfusion and metabolic tracers. In patients with left ventricular dysfunction, the presence of metabolic activity in a region of myocardium supplied by a severely stenotic coronary artery suggests viable tissue that may regain more

normal function after revascularization (Fig. 5-14). PET is less widely available than conventional SPECT imaging; however, PET offers improved spatial resolution because of the higher energy of the isotopes used for this type of imaging.

CARDIAC CATHETERIZATION Cardiac catheterization is an invasive technique in which fluid-filled catheters are introduced percutaneously into the arterial and venous circulation. This method allows for the direct measurement of intracardiac pressures and oxygen saturation and, with the injection of a contrast agent, visualization of the coronary arteries, cardiac chambers, and great vessels. Cardiac catheterization is generally indicated when a clinically suggested cardiac abnormality requires confirmation and its anatomic and physiologic importance needs to be quantified. In the current era, coronary angiography for the diagnosis of CAD is the most common indication for this test. Noninvasive testing, compared with catheterization, is safer, cheaper, and equally effective in the evaluation of most valvular and hemodynamic questions. Most often, catheterization will precede some type of beneficial intervention, such as coronary artery angioplasty, coronary bypass surgery, or valvular surgery. Although cardiac catheterization is generally safe (0.1% to 0.2% overall mortality rate), procedure-related complications such as vascular injury, renal failure, stroke, and MI can occur. An important objective during the cardiac catheterization is to document the filling pressures within the heart and great vessels. This task is accomplished through use of fluidfilled catheters that transmit intracardiac pressures to a transducer that displays the pressure waveform on an oscilloscope. During a right ventricular heart catheterization, pressures within the right atrium, right ventricle, and pulmonary artery are routinely measured in this manner. The

Chapter 5—Diagnostic Tests and Procedures in the Patient with Cardiovascular Disease

[13N]-ammonia

[18F]-deoxyglucose

[13N]-ammonia

[18F]-deoxyglucose

[13N]-ammonia

[18F]-deoxyglucose

Figure 5-14 Resting myocardial perfusion (obtained with [13N]-ammonia) and metabolism (obtained with [18F]-deoxyglucose) Positron-emission tomography images of a patient with ischemic cardiomyopathy. The study demonstrates a perfusion-metabolic mismatch (reflecting hibernating myocardium) in which large areas of hypoperfused (solid arrows) but metabolically viable (open arrows) myocardium are involving the anterior, septal, and inferior walls and the left ventricular apex. See Web Figure 5-5 for a dynamic image obtained with cardiac single-photon emission computed tomography imaging. (Courtesy of Marcelo F. Di Carli, MD, Brigham and Women’s Hospital, Boston.)

catheter can then be advanced further until it wedges in the distal pulmonary artery. The transmitted pressure measured in this location originates from the pulmonary venous system and is known as the pulmonary capillary wedge pressure. In the absence of pulmonary venous disease, the pulmonary capillary wedge pressure reflects left atrial pressure and, similarly, if no significant mitral valve pathologic condition exists, reflects left ventricular diastolic pressure. A more direct method of obtaining left ventricular filling pressures is to advance an arterial catheter into the left ventricular cavity. With these two methods of obtaining intracardiac pressures, each chamber of the heart can be assessed and the gradients across any of the valves determined (Fig. 5-15). Cardiac output can be determined by one of two widely accepted methods: the Fick oxygen method and the indicator dilution technique. The basis of the Fick method

is that total uptake or release of a substance by an organ is equal to the product of blood flow to that organ and the concentration difference of that substance between the arterial and venous circulation of that organ. If this method is applied to the lungs, the substance released into the blood is oxygen; if no intrapulmonary shunts exist, pulmonary blood flow is equal to systemic blood flow or cardiac output. Thus the cardiac output can be determined by the following equation: Cardiac output = oxygen consumption (arterial oxygen content − venous oxygen content ) Oxygen consumption is measured in milliliters per minute by collecting the patient’s expired air over a known period while simultaneously measuring oxygen saturation in a

Section III—Cardiovascular Disease

mm Hg

200

100

Figure 5-15 Electrocardiographic and left ventricular (LV) and aortic (AO) pressure curves in a patient with aortic stenosis. A pressure gradient occurs across the aortic valve during systole.

sample of arterial and mixed venous blood (arterial and venous oxygen content, respectively, measured in milliliters per liter). The cardiac output is expressed in liters per minute and then corrected for body surface area (cardiac index). The normal range of cardiac index is 2.6 to 4.2 L/min/m2. Cardiac output can also be determined by the indicator dilution technique, which most commonly uses cold saline as the indicator. With this method, cold saline is injected into the blood, and the resulting temperature change downstream is monitored. This action generates a curve in which temperature change is plotted over time, and the area under the curve represents cardiac output. Detection and localization of intracardiac shunts can be performed by sequential measurement of oxygen saturation in the venous system, right side of the heart, and two main pulmonary arteries. In patients with left-to-right shunt flow, an increase in the oxygen saturation, or step-up, will occur as one sample from the chamber where arterial blood is mixing with venous blood. By using the Fick method for calculating blood flow in the pulmonary and systemic systems, the shunt ratio can be calculated. Noninvasive approaches have large supplanted catheterization laboratory assessment of shunts. Left ventricular size, wall motion, and ejection fraction can be accurately assessed by injecting contrast into the left ventricle (left ventriculography). Aortic and mitral valve insufficiency can be qualitatively assessed during angiography by observing the reflux of contrast medium into the left ventricle and left atrium, respectively. The degree of valvular stenosis can be determined by measuring pressure gradients across the valve and determining cardiac output (Gorlin formula). The coronary anatomy can be defined by injecting contrast medium into the coronary tree. Atherosclerotic lesions appear as narrowing of the internal diameter (lumen) of the vessel. A hemodynamically important stenosis is defined as 70% or more narrowing of the luminal diameter. However, the hemodynamic significance of a lesion can be underesti-

mated by coronary angiography, particularly in settings in which the atherosclerotic plaque is eccentric or elongated. Use of intravascular ultrasound, Doppler flow wires, or miniaturized pressure sensors can be used during invasive procedures to help evaluate the severity or estimate the physiologic significance of intermediate lesions. Biopsy of the ventricular endomyocardium can be performed during cardiac catheterization. With this technique, a bioptome is introduced into the venous system through the right internal jugular vein and guided into the right ventricle by fluoroscopy. Small samples of the endocardium are then taken for histologic evaluation. The primary indication for endomyocardial biopsy is the diagnosis of rejection after cardiac transplantation and documentation of cardiac amyloidosis; however, endomyocardial biopsy may have some use in diagnosing specific etiologic agents responsible for myocarditis.

RIGHT VENTRICULAR HEART CATHETERIZATION A right ventricular heart catheterization can be performed at the bedside with a balloon-tipped pulmonary artery (Swan-Ganz) catheter. This technique allows for serial measurements of right atrial, pulmonary artery, and pulmonary capillary wedge pressures as well as cardiac output by thermodilution (Fig. 5-16). Such measurements may be useful in monitoring the response to various treatments, such as diuretic therapy, inotropic agents, and vasopressors (Table 5-4). The pulmonary artery catheter is most useful in the critically ill patient for assessing volume status and differentiating cardiogenic from noncardiogenic pulmonary edema. Notably, however, several papers have suggested no improvements in outcomes of critically ill patients in whom pulmonary artery catheterization was performed. Improvements in noninvasive imaging techniques have made the pulmonary artery catheter much less important in diag nosing cardiac conditions, such as pericardial tamponade, constrictive pericarditis, right ventricular infarction, and ventricular septal defect.

MAGNETIC RESONANCE IMAGING Magnetic resonance angiography or imaging (MRI) is an increasingly used noninvasive method for studying the heart and vasculature, especially in patients who have contraindications to standard contrast angiography. MRI offers highresolution dynamic and static images of the heart that can be obtained in any plane. Good-quality images can be obtained in a higher number of subjects than is typically possible with echocardiography. Obesity, claustrophobia, inability to perform multiple breath-holds of 10 to 20 seconds, and arrhythmias are all causes of reduced image quality. Currently, the presence of cardiac pacemakers or implantable defibrillators is considered a contraindication for MRI. Magnetic resonance angiography is useful in the evaluation of cerebral, renovascular, and lower extremity arterial disease. MRI offers significant advantages over other imaging techniques for the characterization of different tissues (e.g., muscle, fat, scar). MRI is useful in the evaluation of ischemic heart disease because stress-rest myocardial perfusion (Fig. 5-17A) and areas of prior infarction

Chapter 5—Diagnostic Tests and Procedures in the Patient with Cardiovascular Disease

ECG

B Radial artery pressure (mm Hg)

100

C Pulmonary capillary wedge pressure (mm Hg)

D Right atrial pressure (mm Hg)

0 Figure 5-16 Electrocardiographic (ECG) and Swan-Ganz flotation catheter recordings are shown in tracings A and C, respectively. The left portion of tracing C was obtained with the balloon inflated, yielding the pulmonary arterial wedge pressure. The right portion of tracing C was recorded with the balloon deflated, depicting the pulmonary arterial pressure. In this patient, the pulmonary arterial wedge pressure (left ventricular filling pressure) is normal, and the pulmonary artery pressure is elevated because of lung disease.

(see Fig. 5-17B to D) can be visualized with excellent special resolution. The presence of delayed gadolinium contrast enhancement within the myocardium is characteristic of scar or permanently damaged tissue (Web Fig. 5-6). The greater the transmural extent of delayed enhancement in a given segment, the lower is the likelihood of improved function in that segment after revascularization. Because of the better spatial resolution, delayed enhancement imaging can depict localized or subendocardial scars that are not detectable with nuclear imaging techniques. The combined use of stress-rest perfusion and delayed enhancement imaging

has performance characteristics for diagnosing CAD that are at least as good as, and probably superior to, those of conventional stress tests using nuclear scintigraphy or echocardiography. MRI is excellent for evaluating a variety of cardiomyopathies (Fig. 5-18). In addition to morphology and function, characteristic patterns of delayed enhancement have been reported in myocarditis, hypertrophic cardiomyopathy, and cardiac amyloidosis. MRI has also been used to help assess right ventricular morphology and function in patients with suspected arrhythmogenic right ventricular cardiomyopathy.

Section III—Cardiovascular Disease

Table 5-4 Differential Diagnosis Using a Bedside Balloon Flow-Directed (Swan-Ganz) Catheter Disease State

Thermodilution Cardiac Output

PCW Pressure

RA Pressure

Comments

Cardiogenic shock Septic shock (early)

↓ ↑

↑ ↓

nl or ↓ ↓

Volume overload Volume depletion Noncardiac pulmonary edema Pulmonary heart disease RV infarction Pericardial tamponade

nl or ↑ ↓ nl

↑ ↓ nl

↑ Systemic vascular resistance ↑ Systemic vascular resistance; myocardial dysfunction can occur late — — —

nl or ↑ ↓ ↓

nl ↓ or nl nl or ↑

↑ ↑ ↑

Papillary muscle rupture Ventricular septal rupture

↓ ↑

↑ ↑

nl or ↑ nl or ↑

↑ PA pressure — Equalization of diastolic RA, RV, PA, and PCW pressure Large v waves in PCW tracing Artifact caused by RA → PA sampling higher in PA than RA; may have large v waves in PCW tracing

↑, Increased; ↓, decreased; nl, normal; PA, pulmonary artery; PCW, pulmonary capillary wedge; RA, right atrium; RV, right ventricle.

Figure 5-17 Cardiac magnetic resonance imaging (MRI) showing use of cardiac MRI in evaluation of cardiomyopathies. A, Severe left ventricular hypertrophy in a patient with hypertrophic cardiomyopathy. Diastolic frame shows open mitral valve. B, Systolic frame showing systolic anterior motion of mitral valve with flow disturbance in left ventricular outflow tract. C, Patient with left ventricular noncompaction as evidenced by deep trabeculations in the left ventricular apex. D, Patient with ischemic cardiomyopathy who has transmural apical infarction and adjacent mural thrombus. See Web Figure 5-6 for a dynamic cardiac MRI image. (Images courtesy of Sheldon E. Litwin, MD, Division of Cardiology, University of Utah.)

COMPUTED TOMOGRAPHY OF THE HEART New applications of computed tomography (CT) have greatly advanced our ability to diagnose cardiovascular disease noninvasively. The development of fast gantry rotation speeds and the addition of multiple rows of detectors

(multidetector CT) has allowed unprecedented visualization of the great vessels, heart, and coronary arteries with images acquired during a single breath-hold (10 to 15 seconds). Until recently, CT has been used most frequently to diagnose aortic aneurysm and acute aortic dissection and pulmonary embolism. CT is also useful for defining con-

Chapter 5—Diagnostic Tests and Procedures in the Patient with Cardiovascular Disease

Figure 5-18 Use of cardiac magnetic resonance imaging in the evaluation of chest pain or ischemic heart disease. A, First-pass perfusion study during vasodilator stress showing large septal perfusion defect. The hypoperfused area appears dark compared with the myocardium with normal perfusion. B, Example of delayed enhancement imaging with nearly transmural infarction of the mid-inferolateral wall, including the posterior papillary muscle. Infarcted myocardium appears white, whereas normal myocardium is black. C, Nontransmural (subendocardial) infarction of the septum and apex. D, Patient with acute myocarditis mimicking an acute coronary syndrome. Mid-myocardial, rather than subendocardial, delayed enhancement is characteristic of myocarditis.

genital abnormalities and detecting pericardial thickening or calcification associated with constrictive pericarditis. More recently, ECG-gated dynamic CT images have been used to quantify ventricular size, function, and regional wall motion (Web Fig. 5-7), and in contrast to echocardiography, CT is not limited by the presence of lung disease or chest wall deformity. However, obesity and the presence of prosthetic materials (i.e., mechanical valves or pacing wires) may affect image quality. The greatest excitement and controversy over cardiac CT relates to the evaluation of coronary atherosclerosis. Electron-beam and multidetector CT scans can be used to quickly and reliably visualize and quantitate the extent of coronary artery calcification (Fig. 5-19). The presence of coronary calcium is pathognomonic of atherosclerosis, and the extent of coronary calcium (usually reported as an Agatston score) is a powerful marker of future cardiac events. The coronary calcium score adds substantial, independent

improvement in risk prediction to the commonly employed clinical risk scores (e.g., the Framingham risk score). Although the extent of coronary artery calcification does not reliably predict the severity of stenoses, the calcium score is a good marker of the overall atherosclerotic burden. Contrast-enhanced coronary computed tomographic angiography (CTA) has improved dramatically in recent years. Coronary CTA has been reported to have a sensitivity of more than 95% in diagnosing significant coronary artery obstruction. This is superior to the sensitivity of stress echo or nuclear myocardial perfusion scanning. Given the speed and accuracy of this test, it is likely to assume a major role in the evaluation of patients with acute chest pain syndromes. Some advocates of cardiac CT have proposed the use of this test for the triple rule-out in patients with acute chest pain—namely, the ability to diagnose pulmonary embolism, aortic dissection, and coronary artery disease with one imaging study. Formal evaluation of this hypoth-

Section III—Cardiovascular Disease

Figure 5-19 Computed tomographic coronary angiography compared with conventional radiographic contrast angiography. A and B, Volume-rendering technique demonstrating stenosis of the right coronary artery and normal left coronary artery. C and D, Maximal intensity projection of the same arteries demonstrating severe noncalcified plaque in the right coronary artery with superficial calcified plaque. E and F, Invasive angiography of the same arteries. (From Raff GL, Gallagher MJ, O’Neill WW, etal: Diagnostic accuracy of noninvasive coronary angiography using 64-slice spiral computed tomography. J Am Coll Cardiol 46:552, 2005.)

esis still needs to be undertaken. Detractors of cardiac CT most frequently cite the risks of radiation and contrast exposure as well as a lack of prospective studies showing improvement in outcome with this testing modality. Of note, the calculated radiation exposure of a cardiac CTA is about double that of a diagnostic invasive coronary angio gram, but is similar to that of a typical nuclear myocardial perfusion scan. As of this writing, the future role of cardiac CTA in routine clinical practice remains uncertain.

NONINVASIVE VASCULAR TESTING Assessment for the presence and severity of peripheral vascular disease is an important component of the cardiovascular evaluation. Comparison of the systolic blood pressure in the upper and lower extremities is one of the simplest tests to detect the presence of hemodynamically important arterial disease. Normally, the systolic pressure in the thigh is similar to that in the brachial artery. An ankle-to-brachial

Chapter 5—Diagnostic Tests and Procedures in the Patient with Cardiovascular Disease pressure ratio (ankle-brachial index) of less than or equal to 0.9 is abnormal. Patients with claudication usually have an index ranging from 0.5 to 0.8, and patients with rest pain have an index less than 0.5. In some patients, measuring the ankle-brachial index after treadmill exercise may be helpful in identifying the importance of borderline lesions. During normal exercise, blood flow increases to the upper and lower extremities and decreases in peripheral vascular resistance, whereas the ankle-brachial index remains unchanged. In the presence of a hemodynamically significant lesion, the increase in systolic blood pressure in the arm is not matched by an increase in blood pressure in the leg. As a result, the ankle-brachial index will decrease, the magnitude of which is proportional to the severity of the stenosis. After significant vascular disease in the extremities has been identified, plethysmography can be used to determine the location and severity of the disease. With this method, a pneumatic cuff is positioned on the leg or thigh and, when inflated, temporarily obstructs venous return. Volume

changes in the limb segment below the cuff are converted to a pressure waveform, which can then be analyzed. The degree of amplitude reduction in the pressure waveform corresponds to the severity of arterial disease at that level. Doppler ultrasound uses reflected sound waves to identify and localize stenotic lesions in the peripheral arteries. This test is particularly useful in patients with severely calcified arteries, in whom pneumatic compression is not possible and ankle-brachial indices are inaccurate. In combination with real-time imaging (duplex imaging), this technique is useful in assessing specific arterial segments and bypass grafts for stenotic or occlusive lesions. Both magnetic resonance angiography and CTA allow high-quality and comprehensive imaging of the entire peripheral arterial circulation in a single study. The threedimensional nature of these studies and their ability to perform extensive postprocessing views (including crosssectional views) of any vessel, even those that are very tortuous, are attractive features of these modalities.

Prospectus for the Future Multidisciplinary teams consisting of cardiologists, cardiac surgeons, vascular surgeons, and radiologists will replace existing and traditional approaches for the evaluation and management of patients with cardiac disease. Such collaboration will foster efficiency and rapid advances for improvements in patient care, education, and research within a seamless, integrated environment. Career opportunities within organiza-

References Cheitlin MD, Armstrong WF, Aurigemma GP, et al: ACC/AHA/ASE 2003 guideline update for the clinical application of echocardiography—summary article: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/ASE Committee to Update the 1997 Guidelines for the Clinical Application of Echocardiography). J Am Soc Echocardiogr 16:1091-1110, 2003. Eagle KA, Berger PB, Calkins H, et al: ACC/AHA guideline update for perioperative cardiovascular evaluation for noncardiac surgery—executive summary: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1996 Guidelines on Perioperative Cardiovascular Evaluation of Noncardiac Surgery). Circulation 105:1257-1267, 2002. Gibbons RJ, Abrams J, Chatterjee K, et al: ACC/AHA 2002 guideline update for the management of patients with chronic stable angina—summary article: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on the Management of Patients with Chronic Stable Angina). Circulation 107:149-158, 2003.

tions with the supporting infrastructure for cardiac imaging will likely realize the promise for patient-oriented, team-based cardiovascular medicine. Tissue enhancement using MRI of either the atria or ventricles will assist practitioners with classification and forecasting of outcomes after interventions such as ablation for atrial fibrillation.

Gibbons RJ, Balady GJ, Bricker JT, et al: ACC/AHA 2002 guideline update for exercise testing—summary article: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1997 Exercise Testing Guidelines). J Am Coll Cardiol 40:1531-1540, 2002. Klein C, Nekolla SG: Assessment of myocardial viability with contrast-enhanced magnetic resonance imaging: Comparison with positron emission tomography. Circulation 105:162-167, 2002. Morey SS: ACC and AHA update guidelines for coronary angiography. American College of Cardiology. American Heart Association. Am Fam Physician 60:1017-1020, 1999. Raff GL, Goldstein JA: Coronary angiography by computed tomography. J Am Coll Cardiol 49:1830-1833, 2007. Sandham JD, Hull RD, Brant RF, et al: A randomized, controlled trial of the use of pulmonary artery catheters in high-risk surgical patients. N Engl J Med 348:5-14, 2003.

III

Chapter

Heart Failure and Cardiomyopathy Sheldon E. Litwin and Ivor J. Benjamin

he syndrome of heart failure occurs when an abnormality of cardiac function results in failure to provide adequate blood flow to meet the metabolic needs of the body’s tissues and organs or in an excessive rise in cardiac filling pressures. In most cases, myocardial dysfunction causes impaired ventricular filling, as well as emptying. Heart failure can result from a large number of heterogeneous disorders (Table 6-1). Idiopathic cardiomyopathy is defined as a primary abnormality of myocardial tissue in the absence of coronary occlusive, valvular, or systemic disease. However, in the clinical setting, the term cardiomyopathy is often used to refer to myocardial dysfunction that is the result of a known genetic, cardiac, or systemic disease. These secondary cardiomyopathies may be related to a significant number of disorders, but in the United States, they are most often the result of ischemic heart disease. Ventricular dysfunction can also result from excessive pressure overload, such as with long-standing hypertension or aortic stenosis, or volume overload, such as aortic insufficiency or mitral regurgitation. Diseases that result in infiltration and replacement of normal myocardial tissue, such as amyloidosis, are rare causes of heart failure. Hemochromatosis can cause a dilated cardiomyopathy that is believed to result from iron-mediated mitochondrial damage. Diseases of the pericardium, such as chronic pericarditis or pericardial tamponade, can impair cardiac function without directly affecting the myocardial tissue. Long-standing tachyarrhythmias have been associated with myocardial dys function that is often reversible. In addition, an individual with underlying myocardial or valvular disease may develop heart failure with the acute onset of an arrhythmia. Finally, multiple metabolic abnormalities (e.g., thiamine deficiency, thyrotoxicosis), drugs (e.g., alcohol, doxorubicin), and toxic chemicals (e.g., lead, cobalt) can damage the myocardium.

Forms of Heart Failure Heart failure can be classified as predominantly left or right sided, high output or low output, and acute or chronic. High-output failure is an uncommon disorder that can occur with severe anemia, vascular shunting, or thyrotoxicosis. This failure results when the heart is unable to meet the abnormally elevated metabolic demands of the peripheral tissues even though cardiac output is elevated. Fluid retention is a common component of this syndrome. Lowoutput failure is much more common than high-output failure and is characterized by insufficient forward output, particularly during times of increased metabolic demand. Cardiac dysfunction may predominantly affect the left ventricle, as with a large myocardial infarction, or the right ventricle, as with an acute pulmonary embolus; however, in many disease states, both ventricles will be impaired (biventricular heart failure). Acute heart failure usually refers to the situation in which an individual who was previously asymptomatic develops heart failure signs or symptoms following an acute injury to the heart, such as myocardial infarction, myocarditis, or rupture of a heart valve. Chronic heart failure refers to the situation in which an individual whose symptoms have developed over a long period, most often when preexisting cardiac disease is present. However, a patient with myocardial dysfunction from any cause may be well compensated for long periods and then develop acute heart failure symptoms in the setting of arrhythmia, anemia, hypertension, ischemia or infection. The severity of heart failure symptoms does not correlate closely with the usual clinical measures of cardiac function (i.e., left ventricular ejection fraction [LVEF]), although the LVEF is a good prognostic marker. This situation likely reflects the fact that ventricular filling pressures are a more important determinant of symptoms than myocardial

Chapter 6—Heart Failure and Cardiomyopathy Table 6-1 Causes of Congestive Heart Failure and Cardiomyopathy Coronary Artery Disease Acute ischemia Myocardial infarction Ischemic cardiomyopathy with hibernating myocardium Idiopathic Idiopathic dilated cardiomyopathy* Idiopathic restrictive cardiomyopathy Peripartum Pressure Overload Hypertension Aortic stenosis Volume Overload Mitral regurgitation Aortic insufficiency Anemia Atrioventricular fistula Toxins Ethanol Cocaine Doxorubicin (Adriamycin) Methamphetamine Metabolic-Endocrine Thiamine deficiency Diabetes Hemochromatosis Thyrotoxicosis Obesity Hemochromatosis Infiltrative Amyloidosis Inflammatory Viral myocarditis Hereditary Hypertrophic Dilated *Genetic bases for these cardiomyopathies have been identified in a large number of individual patients and families. Most of the mutations have been found in cardiac contractile or structural proteins.

function per se. Heart failure may occur in the setting of a reduced or preserved ejection fraction (EF). Recent data suggest that when sensitive methods for assessing myocardial function (i.e., tissue velocity or strain rate imaging) are used, changes are usually detected in both systolic and diastolic function in patients with heart failure (even when the EF is normal or near normal). Importantly, the pre disposing conditions for heart failure (e.g., hypertension, advanced age, coronary artery disease, renal dysfunction) are similar, the prognosis is similar irrespective of whether the LVEF is preserved or reduced. Despite many similarities, medical treatments that have been proved beneficial in heart failure with reduced EF have not shown similar efficacy in heart failure with preserved ejection fraction.

ACUTE PULMONARY EDEMA In patients with the acute onset of pulmonary edema, initial management should be directed at improving oxygenation

and providing hemodynamic stability. These patients commonly have marked elevation of blood pressure, cardiac ischemia, and worsening mitral regurgitation as contributing factors to the pulmonary edema. Standard therapy includes supplemental oxygen and an intravenous loop diuretic. Sublingual or intravenous nitroglycerin helps reduce preload through venodilation and may provide symptomatic relief in patients with ischemic and nonischemic ventricular dysfunction. Intravenous morphine acts in a similar manner but must be used with caution, given its depressive effects on respiratory drive. In patients with hypertensive urgency, severe hypertension, or congestive heart failure related to aortic or mitral regurgitation, an arterial vasodilator, such as nitroprusside, may be helpful in reducing afterload. Evaluation of the patient’s response to treatment requires frequent assessments of blood pressure, heart rate, endorgan perfusion, and oxygen saturation. In patients with persistent hypoxia or respiratory acidosis, mechanical ven tilation or external ventilatory support may be necessary. Pulmonary artery catheterization may be helpful in documenting filling pressures, cardiac output, and peripheral vascular resistance and in monitoring the response to therapy, although invasive monitoring has not been associated with improved patient outcomes. In patients with refractory pulmonary edema or systemic hypotension, an inotropic agent, an intra-aortic balloon pump, or a ventricular assist device may be necessary.

HEART FAILURE WITH PRESERVED EJECTION FRACTION Slowed relaxation of the left ventricle and increased chamber stiffness impair ventricular filling and may contribute to elevated left ventricular, left atrial, and pulmonary venous pressures. Diastolic filling abnormalities contribute to heart failure symptoms in most patients with reduced left ventricular function. However, some patients with a diagnosis of heart failure have normal or nearly normal EF. These patients have been commonly labeled as having diastolic heart failure. As described earlier in this chapter, newer imaging techniques have revealed that most of these patients also have a component of systolic dysfunction as well. Thus, the term heart failure with preserved EF is now the preferred terminology to describe this condition. Relaxation abnormalities are present in most people older than 65 years and are almost universal after age 75 years; however, most of these individuals do not have heart failure. Thus, isolated abnormalities of left ventricular relaxation are apparently not sufficient to directly cause heart failure in the absence of other predisposing conditions. In patients with a variety of cardiovascular diseases, relaxation abnormalities appear at earlier ages than would otherwise be expected. As of this writing, no therapeutic agents that specifically target impaired relaxation have been developed. β–Receptor agonists (dobutamine) and phosphodiesterase inhibitors (milrinone) have potent lusitropic effects (improve relaxation); however, they also directly increase contractility and enhance myocyte calcium cycling. Chronic β-blocker therapy is associated with parallel improvements in systolic and diastolic function, even though both of these may actually deteriorate during the early phases of treatment. Although calcium channel blockers have been

Section III—Cardiovascular Disease

proposed as therapy for diastolic abnormalities, little evidence supports their use for this purpose. Moreover, calcium entry into cardiac myocytes through L-type calcium channels occurs almost exclusively during systole; thus, the theoretical basis for their use is also not firm. In general, all therapies that result in improved systolic function also tend to improve diastolic function, or at least diastolic filling pressures. The use of diuretics to control volume overload and the vigorous treatment of hypertension are the mainstay of therapy for this condition.

RESYNCHRONIZATION THERAPY Interventricular conduction delays, demonstrated as a prolonged QRS duration, are a common complication in patients with heart failure and have been associated with reduced exercise capacity and a poor long-term prognosis. Biventricular pacing or resynchronization therapy results in more normal ventricular contraction and has been associated with an improvement in cardiac output and LVEF. Biventricular pacing may have a beneficial effect on left ventricular remodeling by reducing left ventricular volume, left ventricular mass, and severity of mitral regurgitation. Clinically, these hemodynamic and structural changes have translated into an improvement in exercise duration, functional capacity, and quality of life. Biventricular pacing has also been shown to reduce mortality. Unfortunately, up to 30% of patients undergoing biventricular pacemaker placements do not respond favorably to the treatment. At present, this therapy is generally reserved for patients with severe heart failure and a widened QRS complex who remain symptomatic despite optimal pharmacologic therapy. Intense research efforts are underway to identify with increased accuracy the patients who are likely to derive the greatest benefit. Efforts are currently focused on the quantification of mechanical asynchrony using newer imaging techniques including tissue Doppler imaging and strain imaging (echocardiography or magnetic resonance imaging), wall thickening analysis by computed tomography, and phase imaging with nuclear scintigraphy. Resynchronization therapy is indicated for ambulatory patients with sinus rhythm and class III or IV symptoms. This therapy has not been well studied in patients with atrial fibrillation. Because of the significant expense of this treatment, it is not currently recommended for patients with short life expectancy, including those with refractory, decompensated heart failure.

Adaptive Mechanisms in Heart Failure A large number of compensatory changes occur in the cardiovascular and renal systems to maintain adequate blood flow to the vital organs of the body in the setting of myocardial dysfunction. These changes include increases in left ventricular volume and pressure through the Frank-Starling mechanism, ventricular remodeling, and neurohormonal activation. In the normal heart, increasing the stroke volume or heart rate can augment cardiac output. Stroke volume is dependent on the contractile state of the myocardium, left

Pathophysiology of Heart Failure

Myocardial damage or injury (ischemia, HTN, myocarditis, toxin, etc.)

Neuroendocrine activation (SNS, RAS)

Myocyte hypertrophy, fibrosis, chamber remodeling

Contractility and relaxation Figure 6-1 Schematic diagram illustrating the progressive nature of left ventricular dysfunction that can follow an initial cardiac insult. Attenuation of the neurohumoral activation (or blockade of the downstream effects) may interrupt the positive feedback and slow or reverse the progression of heart failure. HTN, hypertension; RAS, renin-angiotensin system; SNS, sympathetic nervous system.

ventricular filling (preload), and resistance to left ventricular emptying (afterload). According to the Frank-Starling law (Fig. 6-1), stroke volume can be increased with minimal elevation in left ventricular pressure as long as contractility is normal and outflow is not impeded. In the failing heart with depressed intrinsic contractility (Fig. 6-2, curve A), larger increases in filling pressures are required to produce similar increases in stroke volume. When left ventricular diastolic pressure approaches 20 to 25 mm Hg, the hydrostatic pressure in the pulmonary capillaries exceeds the oncotic pressure, and pulmonary edema may ensue. Both depressed myocardial contractility and increased chamber stiffness can lead to pulmonary congestion through similar mechanisms. The failing heart may also undergo changes in left ventricular size, shape, and mass to maintain adequate forward flow. This process is known as remodeling and occurs in response to myocyte loss, such as after a myocardial infarction, or to hemodynamic overload, such as aortic or mitral valve insufficiency. The initial response to increased cardiac stress or load is usually hypertrophy of the viable myocytes. If the increase occurs mainly in cell length, then ventricular dilation is the predominant form of remodeling (usually seen in volume overload or myocardial infarction). The eccentric pattern of remodeling helps maintain cardiac output but occurs at the expense of increased ventricular wall stress. If the myocytes predominantly increase in width (as in the setting of pressure overload), the heart will tend to thicken with maintenance of cavity volume. This form of remodeling, usually referred to as concentric hypertrophy, will tend to reduce wall stress but may do so at the expense of increased filling pressures. If the extent of hypertrophy is inadequate to normalize wall stress, a vicious cycle is established. Overstretching of the myocytes can lead to an increase in myocyte death, ventricular dilation, development of a spherical left ventricular cavity, and further elevation in wall stress.

Chapter 6—Heart Failure and Cardiomyopathy

Normal

Inotropic agent or afterload reduction

Stroke volume

B C Adequate

Table 6-2 New York Heart Association Functional Classification for Heart Failure Class

Patient Symptoms

I (Mild)

No limitation of physical activity. Ordinary physical activity does not cause undue fatigue, palpitation, or dyspnea (shortness of breath). Slight limitation of physical activity. Comfortable at rest, but ordinary physical activity results in fatigue, palpitation, or dyspnea. Marked limitation of physical activity. Comfortable at rest, but less than ordinary activity causes fatigue, palpitation, or dyspnea. Unable to carry out any physical activity without discomfort. Symptoms of cardiac insufficiency at rest. If any physical activity is undertaken, discomfort is increased.

Inadequate A

Depressed contractility

Pulmonary edema 10 20 Left ventricular end-diastolic pressure (mm Hg) Figure 6-2 Normal and abnormal ventricular function curves. When the left ventricular end-diastolic pressure acutely rises above 20mmHg (A), pulmonary edema often occurs. The effect of diuresis or venodilation is to move leftward along the same curve, with a resultant improvement in pulmonary congestion and with minimal decrease in cardiac output. The stroke volume is poor at any point along this depressed contractility curve; thus, therapeutic maneuvers that would raise it more toward the normal curve would be necessary to improve cardiac output significantly. Unlike the effect of diuretics, the effect of digitalis or arterial vasodilator therapy in a patient with heart failure is to move the patient into another ventricular function curve intermediately between the normal and depressed curves. When the patient’s ventricular function moves from A to B by the administration of one of these agents, the left ventricular end-diastolic pressure may also decrease because of improved cardiac function; further administration of diuretics or venodilators may shift the patient further to the left along the same curve from B to C and eliminate the risk for pulmonary edema. A vasodilating agent that has both arteriolar and venous dilating properties (e.g., nitroprusside) would shift this patient directly from A to C. If this agent shifts the patient from A to D because of excessive venodilation or administration of diuretics, then the cardiac output may fall too low, even though the left ventricular end-diastolic pressure would be normal (10mmHg) for a normal heart. Thus, left ventricular end-diastolic pressures between 15 and 18mmHg are usually optimal in the failing heart to maximize cardiac output but avoid pulmonary edema.

The mechanical changes are triggered, in part, by acti vation of several neurohormonal systems. The renin-angio tensin-aldosterone system helps maintain cardiac output through expansion of intravascular volume by promoting retention of sodium and water. Stimulating arterial vasoconstriction through the actions of angiotensin II enhances tissue perfusion. In addition, release of vasopressin will promote free water absorption by the kidney. The sympathetic nervous system helps maintain tissue perfusion by increasing arterial tone as well as increasing heart rate and ventricular contractility. Although adaptive in the short term, activation of these systems is associated with several deleterious effects, including elevation in ventricular filling pressures, depression of stroke volume secondary to an increase in peripheral vascular resistance, and stimulation of myocardial hypertrophy and left ventricular remodeling. These maladaptive changes are ultimately responsible for many of the signs and symptoms associated with congestive heart failure and provide the rationale for treatment.

II (Mild)

III (Moderate)

IV (Severe)

Countering these effects, and in response to the increase in ventricular filling pressures, the myocardial cells secrete atrial natriuretic peptide and brain natriuretic peptide (BNP). The plasma concentration of both of these hormones has been shown to increase in patients with heart failure. The measurement of serum BNP or its precursors has proved to be clinically useful in the diagnosis of heart failure. Although endogenous natriuretic peptides promote salt and water excretion by the kidneys and result in arterial vasodilation, they are relatively ineffective at reversing the maladaptive changes associated with the powerful renin-angiotensin and sympathetic nervous systems.

Evaluation of Patients with Heart Failure The history and physical examination are integral parts of the diagnosis of heart failure and the determination of its underlying or precipitating cause. One of the cardinal manifestations of left ventricular heart failure is dyspnea, which is related to elevation in pulmonary venous pressure. In patients with chronic heart failure, shortness of breath initially occurs only with exertion but may progress to occur at rest. Cardiac dyspnea is often worsened by the recumbent position (orthopnea) when increased venous return further elevates pulmonary venous pressure. Paroxysmal nocturnal dyspnea occurs after several hours of sleep and is probably caused by central redistribution of edema. If cardiac output is low but left ventricular filling pressures are normal, the patient may complain primarily of fatigue resulting from diminished blood flow to the exercising muscles. In some instances, heart failure is slow to progress, and the patient may unknowingly restrict his or her activities. Thus, the history should include an assessment not only of the patient’s symptoms but also of his or her level of activity (functional capacity; Table 6-2). Many patients complain of peripheral

Section III—Cardiovascular Disease

edema, usually involving the lower extremities. The edema commonly worsens during the day and decreases overnight with elevation of the legs. In patients with severe, longstanding heart failure, the edema can involve the thighs and abdomen, and ascites may develop. Importantly, peripheral edema often does not have a cardiac cause. Many of the physical findings of heart failure are related to the neurohormonal changes that help compensate for the reduced cardiac output. An increased heart rate may be present as a result of increased sympathetic tone. The pulse pressure may be narrowed secondary to peripheral vasoconstriction and low stroke volume. If left ventricular filling pressures are elevated, then crackles may be heard on auscultation of the lung fields. Elevation in right-sided filling pressures will result in distended neck veins. If the liver is also congested, firm pressure applied to the right upper quadrant will cause the jugular veins to become further engorged (hepatojugular reflux). Palpation of the precordium may reveal left ventricular enlargement. An earlydiastolic third heart sound (S3) or gallop suggests elevated atrial pressure and increased ventricular chamber stiffness. The sound results from rapid deceleration of the passive component of blood flow from the atrium into the noncompliant ventricle. An S3 can be generated from the left or right ventricle. A fourth heart sound (S4) suggests an increased atrial contribution to left ventricular filling but is not specific for heart failure. The murmurs of both mitral and tricuspid regurgitation are common in patients with congestive heart failure and may become accentuated during an acute decompensation. As stated earlier, peripheral edema is a common finding on physical examination and may be related to elevation in venous pressure or increased sodium and water retention. In bedridden patients, the edema may predominantly be in the presacral region. The electrocardiogram in patients with congestive heart failure is not specific, but it may provide insight into the

cause of the cardiac dysfunction, such as prior myocardial infarction, left ventricular hypertrophy, or significant arrhythmias. The chest radiograph may show chamber enlargement and signs of pulmonary congestion (Fig. 6-3). Treatment of heart failure will result in improvement of the vascular congestion on the chest radiograph, but these changes may lag 24 to 48 hours behind clinical improvement. Certain blood chemistries may be altered in the patient with heart failure. The serum sodium concentration may be low, owing to increased water retention with activation of the renin-angiotensin system. The use of potent diuretics is almost always partially responsible for the hyponatremia. Renal function may be impaired secondary to intrinsic kidney disease or reduced perfusion secondary to renal artery vasoconstriction and low cardiac output. Hepatic congestion is common with right ventricular heart failure and may result in elevated liver enzyme levels. Because many of the signs and symptoms of heart failure may also occur with pulmonary disease, differentiating between these two disease processes may be difficult. Initial therapy will often be directed at both potential pulmonary and cardiac causes until further testing can be performed. Echocardiography is arguably the central means for diagnostic testing in patients with suspected heart failure. This test is fast, safe, and portable and allows for noninvasive assessments of chamber sizes, systolic function, valvular function, both right and left ventricular filling pressures, and quantification of stroke volume or cardiac output. Documentation of heart size, wall thickness, and ventricular function will have important therapeutic implications in most patients (Fig. 6-4). Rapid measurement of the plasma concentration of BNP provides objective and complementary data to aid in the diagnosis of heart failure in the patient with dyspnea. Clinical studies have shown that plasma levels of BNP are elevated in patients with symptomatic left or right ventric ular dysfunction but are usually normal in patients with

Figure 6-3 A, Posteroanterior chest radiograph showing cardiomegaly. B, Lateral chest radiograph showing pulmonary vascular congestion typical of pulmonary edema.

Chapter 6—Heart Failure and Cardiomyopathy

Figure 6-4 Echocardiographic examples of hypertrophic cardiomyopathy seen in long axis (A) and short axis (B) views. Note normal size of left ventricular (LV) cavity and marked thickening of interventricular septum (S) compared with posterior wall (P). In contrast, similar views in a patient with dilated cardiomyopathy (C and D) reveal a markedly enlarged left ventricular cavity with diffuse wall thinning.

dyspnea secondary to noncardiac causes. Unfortunately, a relatively large indeterminate range exists in which the test is not helpful. Advanced age and renal dysfunction also reduce the utility of the test, particularly if the BNP concentration is mildly elevated. An important point to note is that pulmonary edema may also be secondary to noncardiac causes, such as sepsis, certain pulmonary infections, drug toxicity, or neurologic injury. This syndrome, termed adult respiratory distress syndrome, can be differentiated from cardiogenic pulmonary edema by the presence of a low or normal pulmonary capillary wedge pressure. The wedge pressure can be estimated noninvasively using left ventricular filling velocities and mitral annular velocities assessed by conventional and tissue Doppler techniques. Peripheral edema may also occur in disease states other than congestive heart failure. Renal

disease, especially nephrotic syndrome, cirrhosis, and severe venous stasis disease, may be associated with peripheral edema.

Treatment Treatment of congestive heart failure should be directed not only at relieving the patient’s symptoms but also at treating the underlying or precipitating causes (Table 6-3) and preventing progression. Patients should be educated about the importance of compliance with medical therapy as well as dietary salt and fluid restriction. Rhythm disturbances, such as atrial fibrillation, may precipitate congestive heart failure and may require specific therapy. Treatment of active coronary artery disease, hypertension, or valvular disease may

Section III—Cardiovascular Disease

Table 6-3 Precipitants of Heart Failure Dietary (sodium and fluid) indiscretion Noncompliance with medications Development of cardiac arrhythmia Anemia Uncontrolled hypertension Superimposed medical illness (pneumonia, renal dysfunction) New cardiac abnormality (acute ischemia, acute valvular insufficiency)

relieve heart failure symptoms. In addition, correction of concomitant medical problems may help stabilize heart function.

NONPHARMACOLOGIC TREATMENT All patients with heart failure should be instructed to restrict sodium intake to about 2g/day. Fluid intake should also be limited to avoid hyponatremia. Weight reduction in the obese patient helps reduce the workload of the failing heart. Although a growing body of data suggests that a higher body mass index may paradoxically have protective effects in patients with heart failure, each condition is an independent risk factor for increased morbidity and mortality of cardiac disease. A supervised exercise cardiac rehabilitation program can help reduce heart failure symptoms and improve functional capacity in select patients. Although briefly popular, external enhanced counterpulsation (EECP) has fallen out of favor.

PHARMACOLOGIC TREATMENT Diuretics Salt and water retention is common in congestive heart failure secondary to activation of the renin-angiotensinaldosterone system. Diuretics help promote renal excretion of sodium and water and provide rapid relief of pulmonary congestion and peripheral edema. Loop diuretics, such as furosemide, are the preferred agents in the treatment of symptomatic heart failure. In patients who are refractory to high doses of these agents, diuretics that block sodium absorption at different sites within the nephron may be beneficial (i.e., thiazide-type diuretics). Diuretic therapy is currently considered a mainstay in the treatment of patients with heart failure and preserved EF who have elevated left ventricular filling pressures or peripheral edema. Spirono lactone is an aldosterone antagonist with weak diuretic effects that has been shown to reduce hospitalizations for heart failure and cardiac mortality in patients with reduced LVEF and New York Heart Association class III or IV symptoms of heart failure. Notably, diuretic therapy will lower intracardiac filling pressures and thus cardiac output through the FrankStarling mechanism. In most patients, this change is well tolerated. However, in some patients, the reduced cardiac output will result in decreased renal perfusion and a rise in the blood urea nitrogen and creatinine levels.

Vasodilators A large number of vasodilators have been shown to reverse the peripheral vasoconstriction that occurs in congestive heart failure. The most important group of vasodilator agents is the angiotensin-converting enzyme (ACE) inhi bitors. These agents help relieve heart failure symptoms, in part, by blocking production of angiotensin II and reducing afterload. In addition, ACE inhibitors have been shown to reduce mortality in patients with both symptomatic and asymptomatic left ventricular dysfunction. The major side effects of ACE inhibitors include hypotension, hyperkalemia, and azotemia. Cough may occur in about 10% of patients and is related to increased bradykinin levels asso ciated with ACE inhibitor use. Hydralazine in combination with oral nitrates has also been shown to reduce mortality in patients with symptomatic congestive heart failure, although not to the degree of ACE inhibitors. This combination provides an alternative to the patient who is ACE inhibitor intolerant or may require additional therapy for blood pressure control. In addition, recent prospective studies reveal that the combination of hydralazine and nitrates was more beneficial than ACE inhibitors in the African American population. A newer class of agents, the angiotensin II receptor antagonists, prevents the binding of angiotensin II to its receptor. This action has the theoretical advantage of blocking the effects of angiotensin II produced in the bloodstream as well as at the tissue level. In addition, the angiotensin II receptor blockers do not interfere with bradykinin metabolism and therefore are not associated with cough. Several studies comparing ACE inhibitors to angiotensin II blockers suggest that these two classes of agents are equally effective in reducing morbidity and mortality in patients with heart failure. The current guidelines for the management of chronic heart failure, however, recommend that angiotensin II receptor blockers be reserved for patients who are intolerant of ACE inhibitors. ACE inhibitors and angiotensin II receptor blocking agents have both been studied in large randomized trials of patients with heart failure and preserved EF, and they have been found to be ineffective in this large patient group. The negative inotropic effects of the calcium channel blockers and their activation of the sympathetic nervous system make these agents less attractive in the treatment of patients with congestive heart failure. In particular, several studies have shown worsening of heart failure symptoms in patients treated with nifedipine. Other calcium channel blockers, such as diltiazem, have been shown to relieve symptoms and increase functional capacity without a deleterious effect on survival in patients with idiopathic dilated cardiomyopathy. Amlodipine has been studied in patients with both ischemic and nonischemic cardiomyopathy and also has not been associated with an increased cardiac morbidity and mortality. In addition, patients with nonischemic cardiomyopathy treated with amlodipine may have a modest survival benefit. Further studies with these agents are necessary before general recommendations regarding their use in patients with heart failure can be made.

Inotropic Agents Inotropic agents help relieve heart failure symptoms by increasing ventricular contractility. The oldest and most commonly used agent in this class is digoxin, which has been

Chapter 6—Heart Failure and Cardiomyopathy associated with symptomatic improvement in heart failure in patients with systolic dysfunction. However, a recent trial found no significant improvement in survival among patients randomized to digoxin compared with patients treated with placebo. A small reduction in hospitalizations and in death secondary to heart failure was observed, but this was counterbalanced by a slight increase in death secondary to arrhythmias. In general, digoxin therapy should be considered in the patient with left ventricular systolic dysfunction who remains symptomatic after treatment with an ACE inhibitor and a diuretic. No evidence has been found that digoxin should be administered to the patient with asymptomatic left ventricular dysfunction. In addition, digoxin may be harmful in patients with infiltrative cardiomyopathies, such as amyloidosis. Digoxin toxicity results in gastrointestinal, neurologic, and generalized systemic side effects as well as causing a number of tachyarrhythmias and bradyarrhythmias. Several other classes of oral inotropic agents have been evaluated for treatment of congestive heart failure, such as flosequinan, milrinone, vesnarinone, and xamoterol. All these agents have been associated with increased mortality with long-term use. More recently, promising data have emerged involving the calcium-sensitizing agent levosimendan. This class of agents has the theoretical advantage of not increasing calcium fluxes into the myocyte; therefore, these agents should be less arrhythmogenic and have a more favorable energetic effect.

β Blockers As previously discussed, many of the symptoms associated with heart failure are related to activation of several neurohormonal systems, including the sympathetic nervous system. Release of catecholamines may initially help maintain blood pressure and cardiac output but, in the long term, may induce further myocardial injury. To date, long-term use of three different β blockers—metoprolol, bisoprolol, and carvedilol—has been shown in clinical trials to improve LVEF and survival in patients with symptomatic left ventricular dysfunction. Of these agents, carvedilol is unique in that it is also an antioxidant and an α blocker, additional properties that may be beneficial in patients with heart failure. Data from clinical trials comparing the efficacy of metoprolol to carvedilol in patients with heart failure have recently been reported. These data suggest superior effects of carvedilol; however, controversy about the experimental design has limited widespread adoption of a single agent. Therapy with one of the aforementioned β blockers should be strongly considered in all patients who have been stabilized on an ACE inhibitor, digoxin, and a diuretic but remain symptomatic (New York Heart Association classes II to IV). β blockers also appear to be effective in patients who are not taking ACE inhibitors. β-blocker therapy is generally withheld from patients with acutely decompensated heart failure or significant volume overload. Gradual up-titration of the dose improves the ability to tolerate these drugs, which are intrinsically negatively inotropic.

Anticoagulation Thrombosis and thromboemboli occur in patients with left ventricular remodeling and congestive heart failure secondary to stasis of blood, intracardiac thrombi, and atrial

arrhythmias. Although long-term warfarin therapy remains controversial, certain patients may benefit from its use, including patients with chronic atrial fibrillation or flutter, patients with definite mural thrombi noted by echocardiography or ventriculography, and patients in sinus rhythm with LVEF less than 20%. In the general heart failure population, prevention of thromboembolism is roughly balanced by increased bleeding risks. Thus, the routine use of anticoagulation is only recommended in heart failure patients with atrial fibrillation, prior arterial embolic events, or mechanical heart valves and in patients who have had an anterior myocardial infarction in the past 3 months.

Refractory Heart Failure Despite adequate medical therapy, many patients with congestive heart failure fail to have significant reduction in their symptoms. In these instances, therapy with intravenous inotropic agents for 24 to 96 hours, sometimes with hemodynamic monitoring (Swan-Ganz catheter), may be necessary to stabilize the patient. One commonly used agent is dobutamine, which enhances contractility of the heart and reduces peripheral vasoconstriction through stimulation of β2 receptors. Milrinone is an intravenous phosphodies terase inhibitor that has similar effects on contractility and afterload. Administration of these agents often promotes diuresis, especially when given concomitantly with intravenous loop diuretics. In patients with markedly elevated systemic vascular resistance, the use of intravenous vasodilators, such as sodium nitroprusside, can significantly reduce afterload and improve cardiac output. A newly available agent, nesiritide, is a recombinant form of human BNP that has been shown to reduce systemic and pulmonary vascular resistance, increase cardiac output, and promote diuresis comparable to standard inotropic agents and vasodilators. Although nesiritide is less likely to provoke serious dysrhythmias compared with dobutamine, recent data have questioned the safety profile and efficacy of nesiritide. Until further studies document safety, nesiritide is not a first-line agent. If the previously mentioned measures fail to produce a satisfactory diuretic response, dopamine given in doses ranging from 2 to 5mcg/kg per minute may facilitate sodium and water excretion by stimulating renal dopaminergic receptors. If heart failure is accompanied by hypotension, higher doses of dopamine may be necessary. With doses higher than 5 mcg/kg per minute, dopamine can increase heart rate and peripheral vascular resistance through stimulation of β1 and α receptors. Although this dose range of dopamine may help stabilize blood pressure, the increase in afterload may have further deleterious effects on the failing heart. In addition, dopamine may provoke arrhythmias that may lead to further hemodynamic instability. If hypotension persists despite dopamine doses greater than 15mcg/kg per minute, mechanical assist devices, such as the intra-aortic balloon pump, should be considered as a means to stabilize the patient. In patients who cannot be weaned from pharmacologic or mechanical support and in ambulatory patients with severe functional impairment refractory to medical therapy, cardiac transplantation should be considered a means to improve

Section III—Cardiovascular Disease

symptoms and prolong survival (see Chapter 12). Currently, the use of cardiac resynchronization therapy is not a routine consideration in this group of patients who have markedly reduced long-term survival.

Cardiovascular Assist Devices The most commonly used mechanical support device is the intra-aortic balloon pump. This device can be inserted percutaneously through the femoral artery and advanced into the descending thoracic aorta. Inflation of the balloon occurs during diastole such that perfusion pressure in the proximal aorta and coronary arteries is enhanced. Deflation, which occurs just before the onset of systole, greatly reduces aortic impedance and thus significantly reduces afterload. This device is particularly useful in stabilizing patients with severe coronary disease before percutaneous or after surgical revascularization. In addition, this device may provide hemodynamic support in patients with severe mitral regurgitation

or acquired ventricular septal defect before surgical repair. In patients with refractory congestive heart failure, the intraaortic balloon pump may serve as a temporizing measure until cardiac transplantation can be performed. In addition to the intra-aortic balloon pump, several ventricular assist devices are available that provide hemodynamic support. These devices can be placed percutaneously but most commonly are implanted through a sternotomy incision. They can be used to support either ventricle. Blood is collected from the right atrium, left atrium, or the left ventricular apex into an extracorporeal reservoir and then actively pumped back into the pulmonary or systemic circulation by the assist device. These units were initially intended to provide hemodynamic support for several days to weeks (most often a bridge to transplantation in the patient who is critically ill). Because of the success with these devices and the limited availability of donor hearts, assist devices are now being implanted as destination therapy. The pump is placed within the peritoneum, and portable battery packs allow the patient to ambulate. Newer left ventricular assist devices, as well as total artificial hearts, are undergoing clinical investigation as permanent cardiac replacement therapy.

Prospectus for the Future External Containment Devices In patients with left ventricular dysfunction, cardiac remodel ing characterized by progressive ventricular chamber dilation and wall thinning can lead to elevation in wall stress and activation of neurohormonal mechanisms that further impair myocardial function. Experimental devices that passively contain the heart have been shown, in animal models, to reduce ventricular cavity size and improve myocardial responsiveness to β-adrenergic stimulation without impairing left ventricular filling or interfering with coronary blood flow. Randomized trials evaluating the efficacy of these devices in patients with end-stage cardiomyopathy are underway.

Mitral Valve Repair Mitral regurgitation contributes to the progression of heart failure in a large number of patients. However, most patients

References Burkhoff D, Maurer MS, Packer M, et al: Heart failure with a normal ejection fraction: Is it really a disorder of diastolic function? Circulation 107:656-658, 2003. Cleland JG, Daubert JD, Erdmann E, et al: The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 352:1539-1549, 2005. Dokainish H, Zoghbi WA, Lakkis NM, et al: Optimal noninvasive assessment of left ventricular filling pressures: A comparison of tissue Doppler echocardiography and B-type natriuretic peptide in patients with pulmonary artery catheters. Circulation 109:2432-2439, 2004.

with severe heart failure are deemed poor surgical candidates. Several new and exciting percutaneous approaches to mitral and aortic valve repair are being explored as ways to treat some of these patients.

Cell-Based Therapies Permanent loss of myocytes is the final common pathway in most forms of heart failure. Presently, some experimental data support the notion that implantation of cells into the failing heart might effectively regenerate new cardiac muscle. Skeletal myoblasts, bone marrow–derived progenitor cells, and embryonic stem cells are all undergoing testing in both animals and humans. Although many types of transplanted cells are able to contract, ineffective formation of gap junctions and hence ineffective electrical continuity between the transplanted cells and the existing myocardial syncytium continue to be obstacles.

Poole-Wilson PA, Swedberg K, Cleland JG, et al: Comparison of carvedilol and metoprolol on clinical outcomes in patients with chronic heart failure in the Carvedilol or Metoprolol European Trial (COMET): Randomised controlled trial. Lancet 362:7-13, 2003. Rose EA, Gelijns AC, Moskowitz AJ, et al: Long-term mechanical left ventricular assistance for end-stage heart failure. N Engl J Med 345:1435-1443, 2001. Taylor AL, Ziesche S, Yancy C, et al: Combination of isosorbide dinitrate and hydralazine in blacks with heart failure. N Engl J Med 351:2049-2057, 2004. Young JB, Abraham WT, Smith AL, et al: Combined cardiac resynchronization and implantable cardioversion defibrillation in advanced chronic heart failure: The MIRACLE ICD trial. JAMA 289:2685-2694, 2003.

Chapter

III

Congenital Heart Disease Kevin J. Whitehead

bout 0.8% of all live births are complicated by congenital cardiac abnormalities, not including infants with bicuspid aortic valve and mitral valve prolapse (see Chapter 8), which are more prevalent (2% and 2.4%, respectively). Congenital heart disease is a major cause of infant morbidity and mortality. As a result of advances in pediatric cardiology and cardiothoracic surgery, about 85% of infants born with congenital heart disease can be expected to survive into adulthood. In turn, adults with congenital heart disease represent a large and growing population that is encountered more frequently in clinical practice. An equal number of adults and children live with congenital heart disease, with an estimated 800,000 adult patients in the United States alone. Most cases of congenital heart disease occur sporadically, without a known specific cause. Genetic abnormalities are responsible for a proportion of cases and may contribute to cases occurring sporadically as well. Environmental factors are also known to cause congenital heart disease. An increased incidence is found in children of patients with congenital heart disease, with a higher risk in mothers than in fathers. In most cases, the nature of the parent’s defect does not predict the lesion in affected offspring. The size and nature of the congenital defect often determine the onset of symptoms. Normal physiologic changes in cardiovascular hemodynamics at birth can prompt pre sentation. Symptoms can develop shortly after birth when transition from fetal to adult circulation represents a new dependence on biventricular circulation with a pulmonary circuit. The isolated pulmonary and systemic circulations of D-transposition of the great arteries become apparent on closure of the last fetal connections between circuits, the ductus arteriosus and the foramen ovale. In other conditions, the primary lesion results in changes that delay presentation until such compensatory mechanisms fail. Hypertrophy of the morphologic right ventricle in L-transposition of the great arteries is sufficient to com pensate for systemic vascular resistance and maintain normal perfusion for years, with symptoms often developing when the systemic ventricle fails. Still other lesions may develop in adulthood when degenerative changes, such as stenosis

of a previously well-functioning bicuspid aortic valve, are superimposed on an initial lesion. Some congenital defects may go undetected throughout life (e.g., small atrial septal defects [ASDs], whereas some may resolve spontaneously (small muscular ventricular septal defects [VSDs]). Many adult patients with congenital heart disease will have already undergone palliative or reparative surgical procedures and will have subsequent care directed at residual defects and sequelae of such procedures. This chapter focuses on the most common congenital abnormalities observed in adults, including those that develop in adulthood and those for which surgical correction during infancy and childhood permits survival into adulthood.

Septal Defects ATRIAL SEPTAL DEFECTS ASDs are some of the most common congenital defects, representing 10% to 17% of cases, with a higher prevalence in women (60%). Defects are classified according to their location in the interatrial septum. The most common ASD (60%), the ostium secundum defect, involves the fossa ovalis. Ostium primum defects (20%) involve the atrioventricular junction and are at one end of the spectrum of atrioventricular septal defects (or endocardial cushion defects). Primum ASDs are usually associated with a cleft mitral valve and mitral regurgitation. In rare cases, primum ASD can be associated with a large VSD and a single atrioventricular (AV) valve, forming an AV septal defect. Sinus venosus defects are located in the superior septum and may be associated with partially anomalous pulmonary venous drainage into the superior vena cava or right atrium. In patients with uncomplicated ASDs (e.g., with normal pulmonary vascular resistance), oxygenated blood shunts from the left to the right atrium. The magnitude of the shunting is determined by the size of the defect and the compliance of the left and right ventricles. Small ASDs accommodate the increased blood flow in the right atrium without sequelae and no significant hemodynamic compro75

Section III—Cardiovascular Disease

mise of the right heart. If the defect is large, the right atrium and right ventricle dilate to accommodate the increased volume of shunted blood (Fig. 7-1). Pressure in the pulmonary artery increases secondary to the increased volume of blood; however, with the exception of extremely large, long-standing defects, pulmonary vascular resistance usually remains normal.

Atrial septal defect

Ventricular septal defect

LV AO

Patent ductus arteriosus RA

Figure 7-1 Diagram illustrates the three types of shunt lesions that commonly survive until adulthood and their effects on chamber size. A, Uncomplicated atrial septal defect demonstrating left-to-right shunt flow across the interatrial septum and resulting in dilation of the right atrium (RA), right ventricle (RV), and pulmonary artery (PA). B, Uncomplicated ventricular septal defect, resulting in dilation of the RV, left atrium (LA), and left ventricle (LV). C, Uncomplicated patent ductus arteriosus, resulting in dilation of the LA, LV, and PA. Ao, aorta. (From Liberthson RR, Waldman H: Congenital heart disease in the adult. In Kloner RA [ed]: The Guide to Cardiology, 3rd ed. Greenwich, Conn, Le Jacq Communications, 1991, pp 24-47. Copyright ©1991 by Le Jacq Communications, Inc.)

Most patients with ASD are asymptomatic until adulthood, when symptoms such as fatigue, dyspnea, and poor exercise tolerance develop, secondary to right ventricular dysfunction. Older patients may decompensate when acquired heart disease leads to a rise in left ventricular filling pressures and more blood is shunted from the left atrium to the already volume-overloaded right heart. Patients with ASD are prone to atrial fibrillation, especially after 50 years of age. Irreversible pulmonary vascular obstruction resulting in right-to-left shunting and cyanosis (Eisenmenger syndrome) is uncommon and occurs infrequently (180 milliseconds) on the surface electro cardiographic study is a marker for increased risk for ventricular tachycardia and sudden death. Palliative surgery may have been performed in childhood to improve pulmonary blood flow. Occasionally, patients may elect not to undergo complete repair. Such palliation involves the creation of a shunt between the systemic and pulmonary circulation (e.g., subclavian artery to ipsilateral pulmonary artery [Blalock-Taussig shunt]), which results in increased pulmonary blood flow and improved oxygenation of the systemic blood. A variety of palliative shunts have

been used for this purpose. Although such procedures often result in long-term palliation of hypoxia, several complications can occur. Patients may outgrow their shunts, or the shunts may spontaneously close and may lead to progressive cyanosis. If the shunt is too large, the increased volume of blood into the pulmonary circulation and left heart may result in pulmonary congestion and progress to irreversible pulmonary vascular obstruction. In patients surviving to adulthood, corrective surgery should still be undertaken, but the operative risk is higher secondary to the presence of right ventricular dysfunction.

COMPLETE TRANSPOSITION OF THE GREAT ARTERIES Complete transposition (also known as D-transposition) represents 5% to 7% of congenital heart disease and is the most common cyanotic congenital heart disease in the newborn. It is characterized by abnormal ventriculoarterial connections with the aorta arising from the right ventricle and the pulmonary artery arising from the left. The circulation is thus two circuits in parallel. This anatomy can support fetal development, but serious consequences result on closure of the foramen ovale and ductus arteriosus shortly after birth, at which point the systemic and pulmonary circuits are separated and oxygenated blood no longer mixes with the systemic circulation. Uncorrected D-transposition has a 90% mortality rate in the first year of life. Associated defects include VSD, left ventricular outflow tract (subpulmonic) stenosis, and coarctation of the aorta. The first successful palliative procedure for D-transposition was the atrial switch procedure (e.g., Mustard or Senning procedures) in which the venous return is baffled to the contralateral ventricle to achieve two circuits in series. These procedures result in excellent short- and mid-term outcomes. Complications include failure of the systemic right ventricle, tricuspid regurgitation, sinus node dysfunction, tachyarrhythmias, and baffle leaks or obstruction. Progressive ventricular failure should prompt consideration of heart transplantation. In the 1980s, the arterial switch procedure supplanted the Mustard and Senning procedures. This technically challenging procedure restores normal anatomy by attaching the aorta to the left ventricle and the pulmonary artery to the right ventricle with reimplantation of the coronary arteries into the new aorta. Less long-term follow-up data are available for patients following arterial switch, but similar favorable mid-term outcomes have been noted, relative to the Mustard and Senning procedures. Late complications include obstruction of the reimplanted coronary arteries with asso ciated myocardial ischemia, and progressive enlargement of the new aortic root with associated valvular regurgitation. It is recommended that all adults be evaluated at least once for coronary artery patency following arterial switch.

CORRECTED TRANSPOSITION OF THE GREAT ARTERIES Inversion of the ventricles and abnormal positioning of the great arteries characterize congenital-corrected transposition of the great arteries (L-transposition). In this anomaly, the anatomic right ventricle lies on the left and receives oxygenated blood from the left atrium. Blood is ejected into

Section III—Cardiovascular Disease

an anteriorly displaced aorta. The anatomic left ventricle lies on the right and receives venous blood from the right atrium and ejects it into the posteriorly displaced pulmonary artery. This condition is not generally cyanotic and is uncommon, representing 0.5% of cases in patients with congenital heart disease. The clinical course of patients with corrected transposition depends on the severity of other intracardiac anomalies. When the abnormality is an isolated lesion, many individuals survive into adulthood without symptoms. In some patients, the systemic ventricle (anatomic right ventricle) may fail, and pulmonary congestion may result. Associated anomalies include atrioventricular nodal block, VSD, and Ebstein anomaly.

SINGLE VENTRICLE AND FONTAN OPERATION A variety of anatomic defects can functionally result in a single ventricle supporting both the pulmonary and systemic circulation. As such, tricuspid atresia, double-inlet left ventricle with VSD, and large atrioventricular septal defect (among others) may all have similar consequences for the patient. The cardiac output is directed in common to both the aorta and the pulmonary artery, with the balance between the two circulatory beds determined by the degree of outflow tract obstruction. If outflow obstruction is equal, the lower pulmonary vascular resistance will tend to favor pulmonary flow; thus, the ideal single ventricle will have some degree of pulmonary outflow obstruction to prevent the development of fixed pulmonary hypertension. Patients with univentricular hearts who are not repaired have a poor prognosis, with a median survival of 14 years of age. Most patients have cyanosis and functional limitations and would benefit from palliative surgery. The goal with palliation is to optimize pulmonary blood flow without volume loading the ventricle. In suitable patients, the Fontan procedure can offer improved functional status and relieve cyanosis. The Fontan procedure and its modifications connect all systemic venous return to the pulmonary artery without an intervening ventricular pump. This can be accomplished by anastomosis of the right atrium to the pulmonary arteries, separate connections between the superior vena cava and the adjacent right pulmonary artery, and the inferior vena cava through a graft to the left pul monary artery or a tunnel connecting the vena cava and anastomosed to the pulmonary artery. The Fontan procedure separates the two circulations and provides relief of cyanosis without providing a volume load on the left ventricle or a pressure load on the pulmonary arteries. Complications include thrombosis, obstruction, or leaks in the Fontan circuit, ventricular dysfunction, arrhythmias, hepatic dysfunction, and protein-losing enteropathy. Patients with poor ventricular function or intractable protein-losing enter opathy after the Fontan procedure should be considered for transplantation.

EISENMENGER SYNDROME In 1897, Victor Eisenmenger first described the clinical and pathologic features of a patient with fixed pulmonary hypertension resulting from a large VSD. In 1958, Paul Wood used

the term Eisenmenger complex to describe the combination of a large VSD with systemic pulmonary pressures and a reversed or bidirectional shunt. The same pulmonary pathologic changes can result from a large shunt at any level, and the term Eisenmenger syndrome was suggested to describe pulmonary hypertension with reversed or bidirectional shunting at any level. Thus a VSD, a PDA, or an ASD could all result in Eisenmenger physiologic characteristics. The defect size generally exceeds 1.5cm in diameter for VSD, with about half that diameter for PDA and twice that diameter for ASD. Large surgical shunts can also lead to Eisenmenger syndrome. Most patients with Eisenmenger syndrome survive to adulthood, with complications generally occurring from the third decade onward. The prognosis is better than that for patients with other causes of pulmonary hypertension such as primary pulmonary hypertension. Complications include hyperviscosity syndrome, hemorrhage or thrombosis, arrhythmias and sudden death, endocarditis and cerebral abscess, ventricular dysfunction, hyperuricemia and gout, and renal impairment, among others. Hyperviscosity syndrome results from excessive erythrocytosis driven by increased erythropoietin in response to chronic hypoxia. Symptoms include headache, myalgias, and altered mentation. Many patients can tolerate a high hematocrit level with mild or no symptoms, and phlebotomy should not be undertaken simply in response to the hematocrit level. Excessive phlebotomy can lead to iron deficiency. Iron-deficient erythrocytes are less distensible and result in higher blood viscosity for any given hematocrit level, with microcytosis being the strongest independent predictor for cerebrovascular events. Patients with Eisenmenger syndrome have achieved a delicate balance, and management of such patients should respect that balance. Prevention of complications is the preferred strategy. Influenza inoculations, endocarditis prophylaxis, and an avoidance of inappropriate phlebotomy are the mainstays of management. Extreme caution should be exercised with noncardiac surgery to avoid precipitous changes in vascular resistance that may lead to cardiovascular collapse. Pregnancy is strongly discouraged, considering the high risk to both the mother and the fetus. Sterilization is preferred because oral contraceptives can aggravate the risk for thrombosis. Pulmonary vasodilator therapy may improve quality of life, but carries recognized risks and should be supervised by expert care.

Other Conditions Congenital anomalies of the coronary arteries are not uncommon and may be asymptomatic or associated with myocardial ischemia. The left circumflex or left anterior descending artery may arise from the right sinus of Valsalva and is usually not associated with abnormalities of myocardial perfusion. Either coronary artery may arise from the right sinus and may pass between the pulmonary trunk and aorta. This abnormality may result in myocardial ischemia, infarction, or sudden death in young adults, especially during exertion. Coronary artery fistulas with drainage into the right ven tricle, vena cava, or pulmonary vein may be associated with myocardial ischemia if a significant amount of coronary blood flow is shunted into the venous system. Diagnosis of these abnormalities is made by coronary angiography.

Chapter 7—Congenital Heart Disease

Prospectus for the Future The growing population of patients with successful outcomes after intervention for congenital heart disease is posing new challenges during adulthood. The increasing prevalence of genetic studies must be linked to genetic counseling. Early prenatal diagnoses that are linked to specific molecular defects

References Deanfield J, Thaulow E, Warnes C, et al: Management of grown up congenital heart disease. Eur Heart J 24:1035, 2003. Gatzoulis MA, Webb GD, Daubeney PEF: Diagnosis and Management of Adult Congenital Heart Disease. Philadelphia, Elsevier, 2003. Therrien J, Dore A, Gersony W, et al: Canadian Cardiovascular Society Consensus Conference 2001 update: Recommendations for the management of adults with congenital heart disease—Part I. Can J Cardiol 17:940, 2001. Therrien J, Gatzoulis M, Graham T, et al: Canadian Cardiovascular Society Consensus Conference 2001 update: Recommendations for the management of adults with congenital heart disease—Part II. Can J Cardiol 17:1029, 2001.

will engender therapeutic strategies in utero. Genetic epidemiologic studies will provide important insights about the influences of the in utero environment on the subsequent susceptibility and risk factors for heart disease during adulthood.

Therrien J, Warnes C, Daliento L, et al: Canadian Cardiovascular Society Consensus Conference 2001 update: Recommendations for the management of adults with congenital heart disease—Part III. Can J Cardiol 17:1135, 2001. Warnes CA, Williams RG, Bashore TM, et al: ACC/AHA 2008 guidelines for the management of adults with congenital heart disease. Circulation 118:e714-e833, 2008. Webb GD, Williams RG: Care of the adult with congenital heart disease. J Am Coll Cardiol 37:1166, 2001. Wilson W, Taubert KA, Gewitz M, et al: Prevention of endocarditis: Guidelines from the American Heart Association. Circulation 116:1736-1754, 2007.

III

Chapter

Acquired Valvular Heart Disease Sheldon E. Litwin

Aortic Stenosis Aortic stenosis can be congenital or acquired in origin (Table 8-1). The most common congenital cardiac abnormality affects the bicuspid aortic valve. Significant narrowing of the orifice usually occurs during middle age after years of turbulent flow through the valve results in leaflet injury, thickening, and calcification. Rheumatic aortic stenosis results from fusion of the leaflet commissures and is usually associated with mitral valve disease. The most common cause of aortic stenosis in adults is degenerative or senile aortic stenosis, which usually occurs in patients older than 65 years. Aortic stenosis is more common in men than it is in women. In patients with aortic stenosis, the outflow obstruction gradually increases over many years, resulting in left ventricular hypertrophy. This response allows the left ventricle to generate and maintain a large pressure gradient across the valve without a reduction in stroke volume. However, left ventricular hypertrophy often results in increased diastolic chamber stiffness because greater intracavitary pressure is required to maintain left ventricular filling. Systolic dysfunction may also occur as a result of changes in expression of myocyte contractile and calcium-cycling proteins. Patients with severe aortic stenosis may be asymptomatic for many years despite the presence of severe obstruction. The cardinal symptoms associated with aortic stenosis are angina, syncope, and congestive heart failure. Angina can occur in the absence of epicardial coronary artery disease because of the increased oxygen demand of the hypertrophied ventricle and decreased coronary blood flow secondary to elevated left ventricular diastolic pressure. Syncope may result from transient arrhythmias but more commonly occurs with exertion when cardiac output is insufficient to maintain arterial pressure in the presence of exercise-induced peripheral vasodilation. Dyspnea may result from increased filling pressures associated with the noncompliant, hypertrophied left ventricle or may signal the onset of systolic 84

dysfunction. Once patients with severe aortic stenosis develop symptoms, the prognosis is poor unless surgical correction is undertaken. Previous studies have shown that the mean survival rate after the onset of symptoms is about 2 years in patients with heart failure, 3 years in patients with syncope, and 5 years in patients with angina (Fig. 8-1). On physical examination, the patient with aortic stenosis may have a laterally displaced, sustained apical impulse secondary to left ventricular hypertrophy (Table 8-2). An audible or palpable S4 may also be present if the patient is in sinus rhythm. Decreased mobility of the aortic cusps may cause the A2 component of S2 to be soft or absent. The murmur of aortic stenosis is a harsh, crescendo-decrescendo murmur that is best heard over the right upper sternal border and often radiates to the neck. As the obstruction increases, the peak of the murmur occurs later in systole. If left ventricular dysfunction develops, the murmur may decrease in intensity secondary to a reduction in stroke volume. The carotid impulse is often diminished in intensity and delayed (i.e., pulsus parvus et tardus) (see Chapter 4), although in older adults, these changes may be present secondary to intrinsic vascular disease in the absence of significant aortic stenosis. The principal electrocardiographic finding in aortic stenosis is left ventricular hypertrophy. Heart block may develop as a result of calcification from the aortic valve extending into the conducting system. Echocardiography is the most important diagnostic test and is useful to determine the cause of the aortic stenosis and to quantitate the degree of obstruction. The mean transvalvular gradient and valve area can be measured and calculated using Doppler techniques. Patients with severe stenosis will often undergo cardiac catheterization both to confirm the presence of severe aortic stenosis and to determine whether concomitant coronary artery disease is present. A valve area less than or equal to 0.7cm2 defines critical aortic stenosis (normal valve area is 3cm2) and is usually associated with a mean transvalvular gradient of more than 50mmHg when normal left

Chapter 8—Acquired Valvular Heart Disease Table 8-1 Major Causes of Valvular Heart Disease in Adults

Aortic Regurgitation Bicuspid aortic valve Aortic dissection Endocarditis Rheumatic fever Aortic root dilation Mitral Stenosis Rheumatic fever

80 60

2 3 5 Average survival years

Average age death (male) 0

A 100

Percent survivors

Chronic Mitral valve prolapse Left ventricular dilation Posterior wall myocardial infarction Rheumatic fever Endocarditis

Tricuspid Regurgitation

Angina Syncope Failure

Latent period (increasing obstruction, myocardial overload)

Mitral Regurgitation

Acute Posterior wall or papillary muscle ischemia Papillary muscle or chordal rupture Endocarditis Prosthetic valve dysfunction Systolic anterior motion of mitral valve

Onset severe symptoms

100 Percent survival

Aortic Stenosis Bicuspid aortic valve Rheumatic fever Degenerative stenosis

60 63 Age, years

AI MI

(35) (70) (124) MS (120) (133) (67) (116) (68) (114) (35) (35) (64) (108) (62) (101) 75 AS(42) (35) (60) (57) (92) (35) (32) (32) (51) (82) (26) (47) (42)

(23)

(42)

(42) (22)

(42)

Functional (annular) dilation Tricuspid valve prolapse Endocarditis Carcinoid heart disease

(76)

(42) (41) (36) (36)

ventricular function is present. It should be noted that in patients with reduced systolic function, the mean gradient might be low despite the presence of severe aortic stenosis. Moreover, symptoms are often present with valve areas of 0.7 to 1cm2. Treatment in most adults with symptomatic aortic stenosis is surgical replacement of the valve. The operative risk and prognosis are best in patients with preserved left ventricular systolic function. However, surgery should still be considered in patients with left ventricular dysfunction because relief of the obstruction can result in significant clinical and hemodynamic improvement. A more nuanced approach is required for patients with asymptomatic aortic stenosis in whom a recommendation for surgical intervention might be based on the functional assessment by exercise stress testing. Advanced age is associated with higher operative morbidity but is not a contraindication to surgical therapy. Balloon aortic valvuloplasty is a percutaneous technique in which a balloon catheter is positioned across the aortic valve. Inflation results in fracture or separation of the fused and calcified cusps. This procedure is most effective in young patients with noncalcified congenital aortic stenosis and is rarely used in adult patients with calcific aortic stenosis because of significant complications and a high restenosis rate (about 30% at 6 months). Medical interventions with cholesterol-lowering therapies to slow the progression of mild to moderate aortic stenosis are still under clinical inves-

6 7 8 4 5 Years since diagnosis

(35) 9

(34) 10

Figure 8-1 A, Natural history of aortic stenosis without surgical therapy. B, Natural history of mitral and aortic valve disease in an era when surgical therapy was not widely available. Survival rates in 42 patients with aortic stenosis (AS, orange circles with dotted line), 35 patients with aortic insufficiency (AI, orange circles with solid line), and 133 patients with mitral insufficiency (MI, yellow circles with solid line). Clinical course in AI, MS (red circles with dotted line), and MI is similar with a 5-year survival rate of about 80% and a 10-year survival rate of about 60%. Patients with AS have a worse prognosis, with 5- and 10-year survival rates of about 40% and 20%, respectively. (A from Ross J Jr, Braunwald E: Aortic stenosis. Circulation 38[Suppl V]:61, 1968. Copyright © 1968 American Heart Association. B from Rapaport E: Natural history of aortic and mitral valve disease, Am J Cardiol 35:221-227, 1975.)

tigation. Routine antibiotic prophylaxis is no longer recommended unless there is prior history of endocarditis.

Aortic Regurgitation Aortic regurgitation (AR) may be secondary to primary disease of the aortic leaflets, aortic root, or both (see Table 8-1). Abnormalities of the aortic leaflets may be secondary

Section III—Cardiovascular Disease

Table 8-2 Characteristic Physical, Electrocardiographic, and Chest Radiographic Findings in Chronic Acquired Valvular Heart Disease Physical Findings* Aortic stenosis

Aortic regurgitation

Mitral stenosis

Mitral regurgitation

Mitral valve prolapse

Tricuspid stenosis

Tricuspid regurgitation

Electrocardiogram

Radiograph

Pulsus parvus et tardus (may be absent in older patients or in patients with associated aortic regurgitation); carotid shudder (coarse thrill) Ejection murmur radiates to base of neck; peaks late in systole if stenosis is severe Sustained but not significantly displaced LV impulse A2 decreased, S2 single or paradoxically split S4 gallop, often palpable Increased pulse pressure Bifid carotid pulses Rapid pulse upstroke and collapse LV impulse hyperdynamic and displaced laterally Diastolic decrescendo murmur; duration related to severity Systolic flow murmur S3G common Loud S1 OS S2-OS interval inversely related to stenosis severity S1 not loud, and OS absent if valve heavily calcified Signs of pulmonary arterial hypertension

LV hypertrophy Left bundle branch block is also common Rare heart block from calcific involvement of conduction system

LV prominence without dilation Post-stenotic aortic root dilation Aortic valve calcification

LV hypertrophy, often with narrow deep Q waves

LV and aortic dilation

Left atrial abnormality Atrial fibrillation common RV hypertrophy pattern may develop if associated pulmonary arterial hypertension is present

Hyperdynamic LV impulse S3 Widely split S2 may occur Holosystolic apical murmur radiating to axilla (murmur may be atypical with acute mitral regurgitation, papillary muscle dysfunction, or mitral valve prolapse) One or more systolic clicks, often mid-systolic, followed by late systolic murmur Auscultatory findings dynamic Symptoms may include tall thin habitus, pectus excavatum, straight back syndrome Jugular venous distention with prominent α wave if sinus rhythm Tricuspid OS and diastolic rumble at left sternal border; may be overshadowed by concomitant mitral stenosis Tricuspid OS and rumble increased during inspiration Jugular venous distention with large regurgitant (systolic) wave Systolic murmur at left sternal border, increased with inspiration Diastolic flow rumble RV S3 increased with inspiration Hepatomegaly with systolic pulsation

LA abnormality LV hypertrophy Atrial fibrillation

Large LA: double-density, posterior displacement of esophagus, elevation of left main stem bronchus Straightening of left heart border as a result of enlarged left appendage Small or normal-sized LV Large pulmonary artery Pulmonary venous congestion Enlarged LA and LV Pulmonary venous congestion

Often normal Occasionally ST-segment depression and/or T-wave changes in inferior leads

Depends on degree of valve regurgitation and presence or absence of those abnormalities

Right atrial abnormality Atrial fibrillation common

Large RA

RA abnormality; findings are often related to cause of the tricuspid regurgitation

RA and RV are enlarged; findings are often related to cause of the tricuspid regurgitation

*Findings are influenced by the severity and chronicity of the valve disorder. LA, left atrium; LV, left ventricle; OS, opening snap; RA, right atrium; RV, right ventricle.

Chapter 8—Acquired Valvular Heart Disease to rheumatic disease, congenital abnormalities, endocarditis, or use of certain anorexigenic drugs. In addition, AR is commonly a consequence of degenerative and bicuspid aortic stenosis. An aortic root pathologic condition associated with annular and root dilation may result in separation or prolapse of the leaflets. With chronic AR, the left ventricle must accommodate the normal inflow from the left atrium in addition to the aortic regurgitant volume. As a result, the left ventricle dilates and hypertrophies to maintain normal effective forward flow and to minimize wall stress. As the AR progresses, these changes in left ventricular size and wall thickness may be insufficient to maintain normal left ventricular filling pressures, and irreversible myocyte damage may occur. As a result, the left ventricle will dilate further, and systolic function and effective stroke volume will decrease. Clinically, patients with chronic, severe AR may be asymptomatic for long periods secondary to the compensatory changes in the left ventricle. When symptoms do develop, they are primarily related to an elevation in left ventricular filling pressures and include dyspnea on exertion, orthopnea, and paroxysmal nocturnal dyspnea. Many patients will describe chest or head pounding secondary to the hyperdynamic circulation. If effective cardiac output is reduced, the patient may complain primarily of fatigue and weakness. As with aortic stenosis, angina may occur in patients with AR even in the absence of epicardial coronary artery disease secondary to elevated left ventricular filling pressures and reduced coronary perfusion pressure. On physical examination, patients with severe AR have a widened pulse pressure (difference between the systolic and diastolic pressures) as a result of the runoff of blood back into the left ventricle (see Table 8-2). The arterial pulse is usually bounding, with a rapid upstroke and quick collapse (Corrigan disease or water-hammer pulse) (see Chapter 4). The cardiac impulse is hyperdynamic and is displaced laterally and inferiorly. The murmur of AR is a high-pitched, decrescendo diastolic murmur best heard at the lower left sternal border with the patient sitting up and leaning forward. Asking the patient to hold his or her breath at end expiration while the hands are held behind the head may also improve the ability to auscultate the murmur of AR. A sys tolic ejection murmur is often heard secondary to increased forward flow across the aortic valve. An S3 gallop may be present, especially if the patient has developed symptoms of heart failure. A low-pitched, diastolic murmur (Austin Flint murmur) may be heard at the apex and confused with the murmur of mitral stenosis (MS). This sound is thought to be secondary to the incomplete opening of the mitral leaflets (functional MS) secondary to elevated left ventricular filling pressures or impingement of the AR jet on the anterior mitral leaflet. The natural history of chronic AR is varied. Many patients with moderate to severe AR will remain asymptomatic for many years and generally have a favorable prognosis. Other patients may have progression of AR severity and develop left ventricular dysfunction and symptoms of congestive heart failure. Echocardiography is the primary tool to monitor the progression of disease and optimize the timing of surgery. Prior studies have shown that patients at high risk are those with left ventricular end-systolic

diameters greater than 50 mm or an ejection fraction of less than 50%. Surgery is usually recommended before developing this degree of left ventricular enlargement or dysfunction. Therefore patients with known moderate to severe AR should be monitored regularly with noninvasive testing to detect early signs of cardiac (i.e., left ventricular) decompensation. Treatment of patients with moderate to severe AR theoretically should include vasodilator therapy, such as nifedipine or angiotensin-converting enzyme (ACE) inhi bitors, because these agents unload the left ventricle. Although some published data suggest that these agents may slow the progression of myocardial dysfunction and delay the need for surgery, more recent data do not support that contention. Valve replacement surgery should be considered in symptomatic patients and those with evidence of significant left ventricular enlargement or left ventricular systolic dysfunction. In patients with reduced left ventricular ejection fraction of short duration (14 months), valve replacement usually results in significant improvement in ventricular function. If left ventricular dysfunction has been present for a prolonged period, then permanent myocardial damage may occur. Although such patients should not be excluded from surgery, their long-term prognosis remains poor. As compared with chronic AR, acute AR is a medical emergency that often requires immediate surgical intervention. The causes of acute AR include infective endocarditis, traumatic rupture of the aortic leaflets, aortic root dissection, and acute dysfunction of a prosthetic valve. Acute AR is the result of hemodynamic instability because the left ventricle is unable to dilate to accommodate the increased diastolic volume, resulting in decreased effective forward flow. Left ventricular and left atrial pressures rise quickly, leading to pulmonary congestion. Patients with acute AR often exhibit symptoms and signs of cardiogenic shock. The patient is usually pale with cool extremities as a result of peripheral vasoconstriction. The pulse is weak and rapid, and the pulse pressure is normal or decreased. The murmur of acute AR is low pitched and short because of rapid equilibration of aortic and left ventricular pressures during diastole. An S3 gallop is often present. Echocardiography is useful to assess AR severity and to determine its cause and can be quickly performed at the bedside in the patient who is acutely ill. The medical treatment of acute AR includes vasodilator therapy and diuretics if the blood pressure is stable. In patients who are hemodynamically compromised, inotropic support and vasopressors may be necessary. For most patients with acute AR, urgent valve replacement remains the treatment of choice. Intra-aortic balloon counterpulsation is relatively contraindicated because it may worsen AR severity.

Mitral Stenosis MS occurs when thickening and immobility of the mitral leaflets impede flow from the left atrium to the left ventricle. Rheumatic fever is by far the most common cause of MS. Rarely, congenital abnormalities, connective tissue disorders, left atrial tumors, and overly aggressive surgical repair

Section III—Cardiovascular Disease Pulmonary congestion threshold

4 cm2

Flow

Mitral valve area 2 cm2

1 cm2

0.5 cm2 Pressure gradient Figure 8-2 Graphic illustration of the relationship between the diastolic gradient across the mitral valve and the flow through the mitral valve. As the mitral valve becomes more stenotic, the pressure gradient across the mitral valve must increase to maintain flow into the left ventricle. When the mitral valve area is 1cm2 or less, the flow rate into the left ventricle cannot be significantly increased, despite a significantly elevated pressure gradient across the mitral valve. (Adapted from Wallace AG: Pathophysiology of cardiovascular disease. In Smith LH Jr, Thier SO [eds]: The International Textbook of Medicine, vol 1. Philadelphia, WB Saunders, 1981, p 1192.)

of a regurgitant valve may lead to obstruction of the mitral valve. Two thirds of patients with MS are women. The pathologic changes that occur with rheumatic MS include fusion of the leaflet commissures and thickening, fibrosis, and calcification of the mitral leaflets and chordae. These changes occur over many years before dysfunction becomes hemodynamically important. The initial hemodynamic change that occurs with MS is an elevated left atrial pressure created by obstruction to left ventricular inflow (Fig. 8-2). This pressure change is transmitted back to the pulmonary venous system and may result in pulmonary congestion. Initially, this change may only occur at more rapid heart rates, such as with exercise or atrial arrhythmias, when higher left atrial pressures develop during the shortened diastolic period. As the MS becomes more severe, left atrial pressure remains elevated even at normal heart rates, and symptoms related to elevated pulmonary venous pressures may be present at rest. Chronic elevations in pulmonary venous pressures may lead to an increase in pulmonary vascular resistance and pulmonary arterial pressures. If the MS is not corrected, irreversible changes in the pulmonary vasculature may occur, and signs and symptoms of right ventricular heart failure may develop. In contrast, left ventricular filling pressures are usually normal or low with mild to moderate MS. As the stenosis becomes severe, filling of the left ventricle is impaired, and stroke volume and cardiac output are reduced. Patients with MS of rheumatic origins usually develop symptoms during the third or fourth decade of life. Dyspnea, orthopnea, and atrial fibrillation are the most common symptoms. Some patients may have sudden hemoptysis secondary to rupture of the dilated bronchial veins (pulmonary apoplexy) or blood-tinged sputum associated with pulmonary edema. Peripheral embolism from left atrial thrombus

may also occur, even in the absence of atrial fibrillation. In long-standing, severe MS, patients may develop peripheral edema secondary to elevated right ventricular pressures and right ventricular dysfunction. Compression of the left recurrent laryngeal nerve from a severely dilated left atrium may result in hoarseness (Ortner syndrome). On physical examination, S1 is loud early in the course of MS because the leaflets remain fully open throughout diastole and then quickly close (see Table 8-2). As the leaflets become more calcified and immobile, S1 will become softer or completely absent. The opening snap is a high-pitched sound after the S2 and reflects the abrupt mitral valve opening. As the MS becomes more severe, the interval between the S2 and opening snap becomes shorter because left atrial pressure exceeds left ventricular pressure earlier in diastole. The characteristic low-pitched rumbling murmur of MS is best heard at the left ventricular apex with the patient in the left lateral decubitus position. The murmur is loudest in early diastole when rapid ventricular filling occurs. If sinus rhythm is present, the murmur may increase in intensity after atrial contraction (presystolic accentuation). In some patients, the murmur may only be heard at times of increased blood flow through the mitral valve, such as after exercise. If pulmonary artery pressures are elevated, a palpable P2 may be detected at the upper left sternal border. On auscultation, the pulmonic component of S2 is prominent and a right ventricular gallop may be present. Echocardiography is the most useful tool for pathologic assessment of the mitral apparatus as well as the severity of the stenosis. The characteristic rheumatic deformity observed with two-dimensional imaging is doming (i.e., hockey stick deformity) of the anterior mitral valve leaflet, which is secondary to fusion of the commissures and tethering of the leaflet tips (Fig. 8-3). In addition, the mobility of the leaflets

LV AMVL

PMVL

Figure 8-3 Example of hockey stick deformity of mitral valve in chronic rheumatic heart disease as visualized by echocardiography. Tips of anterior mitral valve leaflet (AMVL) are tethered, thus restricting opening of the valve. Posterior mitral valve leaflet (PMVL) is thickened and has reduced mobility. Left atrium (LA) is characteristically enlarged.

Chapter 8—Acquired Valvular Heart Disease and the extent of valvular calcification can be assessed and used to determine treatment options. Doppler techniques allow calculation of the mitral valve area and the transvalvular gradient. Transesophageal echocardiography is a useful tool for studying the mitral apparatus and examining the left atrium for thrombus before percutaneous valvuloplasty. The severity of MS and associated hemodynamic changes can also be evaluated with cardiac catheterization. Measurements of the cardiac output and transvalvular gradient can be used to calculate the valve area by means of the Gorlin formula. A normal mitral valve area is 4 to 6cm2, and critical MS is defined as a valve area less than 1cm2. Patients with mild to moderate MS can usually be managed medically. Heart rate control is imperative in these patients because more rapid rates reduce the length of the diastolic filling period. This is especially true in patients with atrial fibrillation, in whom loss of atrial contraction may further reduce left ventricular filling. Anticoagulant therapy is indicated for patients with atrial fibrillation and for those with sinus rhythm who have had prior embolic events or who have moderate to severe MS. Diuretics are useful in relieving pulmonary congestion and signs of right ventricular heart failure. All patients should be instructed on the importance of endocarditis prophylaxis. Prophylaxis against recurrent bouts of rheumatic fever may be used in patients younger than 30 years. Patients with severe symptoms (New York Heart Association classes III through IV) and moderate to severe MS should be considered for a percutaneous or surgical intervention. Percutaneous balloon valvuloplasty is a new technique in which a balloon catheter positioned across the

mitral valve is quickly inflated, resulting in separation of the fused cusps. Optimal short- and long-term results are obtained in patients with pliable, noncalcified leaflets and chords, minimal mitral regurgitation (MR), and no evidence of left atrial thrombus. A surgical option in this same group of patients is open mitral valve commissurotomy. With direct visualization of the mitral valve, the surgeon is able to débride the valve, separate the fused cusps, and remove left atrial thrombi. Although the valve remains abnormal, this procedure is associated with a low operative mortality and a good hemodynamic result and may spare the patient from a valve replacement for many years. If mitral commissurotomy is not an option, valve replacement with a bioprosthetic or mechanical prosthesis can be performed.

Mitral Regurgitation MR can result from abnormalities of the mitral leaflets, annulus, chordae, or papillary muscles (see Table 8-1). The most common leaflet abnormality resulting in chronic MR is myxomatous degeneration of the mitral valves. This condition results in mitral valve prolapse (MVP), which progresses as the chordae become elongated or rupture (Fig. 8-4). Both acute and chronic rheumatic fever may also cause MR. With chronic MR, the left ventricle dilates to compensate for the increased regurgitant volume. However, in contrast to aortic insufficiency, the increased volume is ejected into the low-pressure left atrium. Thus, left ventricular wall stress and pressure remain normal for a significant period. If the left atrium dilates sufficiently to accommodate the increased

AMVL

PMVL

Figure 8-4 Typical example of mitral valve prolapse is visualized with transthoracic echocardiography. A, Prolapse of the posterior mitral valve leaflet (PMVL) behind the mitral valve annular plane results from lengthening and rupture of chordae tendineae. In this patient, a highly eccentric (blue) jet of moderate-to-severe mitral regurgitation is observed (B). AMVL, anterior mitral valve leaflet; LV, left ventricle.

Section III—Cardiovascular Disease

volume, left atrial and pulmonary venous pressures will remain normal. As the MR progresses, myocyte damage may occur, resulting in further left ventricular dilation, an elevation in diastolic filling pressures, and a reduction in left ventricular systolic function. As left atrial and pulmonary venous pressures increase, pulmonary congestion may occur. Patients with chronic compensated MR are usually asymptomatic and have normal functional capacity. When symptoms do occur, left ventricular systolic function is sometimes depressed. Patients may initially complain of fatigue and dyspnea with exertion secondary to reduced cardiac output and elevation in pulmonary venous pressures. If the MR remains untreated, pulmonary hypertension and right ventricular heart failure may occur. MR characteristically produces a holosystolic murmur best heard at the apex and radiating to the axilla and back (see Table 8-2). If an eccentric, anteriorly directed jet of MR is present, an ejection-quality murmur may be present and confused with an aortic outflow murmur. If the MR is secondary to MVP, a mid-systolic click may be present, followed by a late systolic murmur. MR associated with rheumatic mitral disease may be accompanied by heart sounds typical of MS. Echocardiography is the primary noninvasive method for defining mitral valve pathologic evaluation and assessing left ventricular size and function. Doppler techniques are useful in grading the severity of MR. Quantitative echocardiographic measures of MR severity are predictive of long-term survival, even in patients who are asymptomatic. Mitral valve repair appears to normalize the survival curves in this group of patients. MR can also be assessed during cardiac catheterization by estimating the amount of contrast medium that is ejected into the left atrium during left ventriculography. In addition, left ventricular size and systolic function can be quantitated, filling pressures can be measured, and the coronary anatomy can be defined. The medical treatment of patients with compensated chronic MR is afterload reduction with vasodilator therapy, such as ACE inhibitors or hydralazine. The timing of surgery is difficult because the development of symptoms often indicates the presence of left ventricular dysfunction and irreversible myocardial damage. In addition, mitral valve replacement with disruption of the chordal apparatus often results in further left ventricular dilation and decline in systolic function. Echocardiographic parameters that identify patients at risk for a poor response to mitral valve replacement are a left ventricular end-diastolic diameter greater than 70 mm, an end-systolic diameter greater than 45mm, and a low-normal or reduced left ventricular ejection fraction. Patients with known MR should be followed with yearly studies to monitor left ventricular function and size so that surgery can be performed before irreversible myocyte damage and left ventricular remodeling occur. The development of either atrial fibrillation or pulmonary hypertension may be an indication for earlier surgical intervention, even if left ventricle size and function are still normal. In many patients, the mitral valve may be repaired, thus avoiding many of the potential complications associated with valve replacement. With this surgery, sections of redundant leaflet can be excised, leaflets débrided, and chordae

shortened. A prosthetic ring (annuloplasty) can be sewn into the mitral annulus to reduce the size of the orifice and increase the degree of leaflet coaptation. The advantage of this procedure is that preservation of the mitral apparatus helps maintain normal left ventricular geometry and function. In addition, long-term anticoagulation is not necessary in most patients in sinus rhythm. Valve repair is generally not indicated if the mitral valve is heavily calcified or disrupted secondary to papillary muscle disease or endocarditis. In these instances, valve replacement is the procedure of choice. Based on excellent surgical outcomes and long-term durability, mitral valve repair is the procedure of choice in all patients in whom it is technically feasible. Severe MR, even in the absence of symptoms or left ventricular dysfunction, may be an appropriate reason for surgical intervention. A variety of percutaneous mitral valve repair techniques are in development. Acute severe MR is often a life-threatening condition that can result from a variety of papillary muscle, chordal, and leaflet abnormalities (see Table 8-1). Patients with acute MR usually become severely ill because the left atrium does not dilate to accommodate the regurgitant volume. As a result, left atrial and pulmonary venous pressures abruptly increase, resulting in pulmonary congestion. In addition, the decreases in stroke volume and cardiac output result in an increase in systemic vascular resistance and, as a consequence, an increase in the severity of MR. Patients usually exhibit pulmonary edema and signs of cardiogenic shock. On auscultation, the MR murmur is often a soft, low-pitched sound in early systole, resulting from rapid equilibration of left ventricular and left atrial pressures. Afterload reduction with either an intravenous vasodilator, such as nitroprusside, or an intra-aortic balloon pump may help stabilize the patient before urgent valve replacement surgery. Ischemia of the posterior wall or papillary muscles may cause acute but transient MR.

Mitral Valve Prolapse MVP is reported to be present in about 1% to 3% of the population. Although MVP can be observed in all ages and in both sexes, epidemiologic studies suggest that the prevalence is greater in women than it is in men. In some patients, MVP is inherited as an autosomal dominant trait with variable penetrance. MVP is present when superior displacement in ventricular systole of one or both mitral valve leaflets exists across the plane of the mitral annulus toward the left atrium (see Fig. 8-4). Primary or classic MVP occurs when myxomatous degeneration of the mitral valve occurs without evidence of systemic disease. Secondary MVP is also characterized by myxomatous degeneration of the mitral apparatus but in the presence of a recognizable systemic or connective tissue disease, such as Marfan syndrome or systemic lupus erythematosus. Functional MVP results from structural abnormalities of the mitral annulus or papillary muscles or reduced left ventricular volume, but the mitral leaflets are anatomically normal. Most patients with MVP are asymptomatic. Although a variety of nonspecific symptoms have been associated with

Chapter 8—Acquired Valvular Heart Disease

MVP (e.g., chest pain, palpitations, dizziness, anxiety [MVP syndrome]), the frequency of these symptoms is no different from that in the general population. MVP may be associated with varying degrees of MR. MR severity is probably the main determinant of long-term complications. The characteristic physical examination finding in MVP is the midsystolic click, followed by a late systolic murmur (see Table 8-2). The auscultatory findings of MVP are subtle and are greatly affected by changes in left ventricular volume. Maneuvers that reduce left ventricular volume will result in prolapse of the redundant leaflets early in systole; as a result, the click will occur early in systole, and the MR murmur will sound more holosystolic. If left ventricular volume is increased, the click will be heard late in systole, followed by a short systolic murmur. The diagnosis of MVP is usually confirmed by echocardiography, which allows examination of the mitral apparatus and determination of the MR severity. Most patients with mild prolapse and insignificant MR are asymptomatic and require no specific intervention. Endocarditis prophylaxis is generally recommended only if mild or greater MR exists. However, in some individuals, the MR may progress to such a degree that serial examinations and echocardiograms are necessary to monitor MR severity and left ventricular function. Middle-aged and older men and patients with asymmetrical prolapse are at highest risk for developing complications from MVP, such as severe MR and endocarditis. MR that acutely worsens may be related to rupture of the chordae tendineae. Sudden death in the absence of hemodynamically significant MR is rare. Patients with MVP and evidence of structural leaflet abnormalities or significant MR should receive endocarditis prophylaxis. Symptomatic arrhythmias should be treated as discussed in Chapter 10. For patients with severe MR, mitral valve repair or replacement may be indicated as discussed earlier (see “Mitral Regurgitation”).

is a low-pressure system, the mean gradient across the tricuspid valve may be quite small (5mmHg) yet still clinically important.

Tricuspid Stenosis

Pulmonic stenosis is most often congenital in origin and is discussed further in Chapter 7. Rheumatic deformity of the pulmonic valve is rare and not usually associated with hemodynamically important obstruction. Pulmonic regurgitation is most often the result of dilation of the annulus secondary to pulmonary hypertension of any cause. Symptoms are usually related to the primary disease and in most cases are secondary to right ventricular heart failure. In this setting, the murmur of pulmonic regurgitation is a high-pitched, blowing murmur best heard at the second left intercostal space (Graham Steell murmur). In the absence of pulmonary hypertension, the murmur is usually low pitched and occurs late in diastole. Treatment is usually directed at the underlying cause of the pulmonary hypertension. Rarely and usually in the setting of congenital or previously repaired pulmonic valve disease, the valve will need to be replaced because of intractable right ventricular heart failure.

Tricuspid stenosis is most often rheumatic in origin and is usually associated with mitral or aortic disease. Other rare causes include carcinoid syndrome, congenital valve abnormalities, and leaflet tumors or vegetations. Similar to MS, tricuspid stenosis is more common in women than it is in men and tends to be a slowly progressive disease. Patients generally exhibit symptoms and signs of right ventricular heart failure, such as fatigue, abdominal bloating, and peripheral edema. On physical examination, a prominent jugular venous a wave may be present if the patient is in sinus rhythm and may be confused with an arterial pulsation. In addition, a palpable presystolic pulsation coinciding with atrial contraction may be felt on palpation of the liver. On auscultation, the findings of tricuspid stenosis may not be detected secondary to the presence of mitral and aortic valve disease. However, an opening snap may be audible at the left sternal border, followed by a soft, high-pitched diastolic murmur. In contrast to MS, the murmur of tricuspid stenosis is shorter in duration and accentuated with inspiration. Tricuspid stenosis can be diagnosed by echocardiography or right ventricular catheterization. Because the right heart

Tricuspid Regurgitation Tricuspid regurgitation (TR) is most often secondary to dilation of the right ventricle and tricuspid annulus that may occur with right ventricular heart failure of any cause. Other causes include endocarditis, carcinoid syndrome, congenital abnormalities, and chest wall trauma. In the absence of pulmonary hypertension, TR is usually well tolerated. However, if right ventricular dysfunction is present, patients usually have symptoms of right ventricular heart failure. On physical examination, the jugular veins are distended, and a prominent v wave is usually present. Hepatic congestion is common and often associated with a palpable systolic pulsation. The murmur of TR is high pitched and pansystolic and is best heard along the sternal border. Maneuvers that increase venous return, such as inspiration or leg raising, accentuate the murmur and are helpful in differentiating TR from MR or aortic outflow tract murmurs. If the TR is acute, the murmur is usually soft and present only during early systole. TR related to pulmonary hypertension and right ven tricular dysfunction will usually significantly improve with treatment of the underlying cause. Repair of the tricuspid annulus (annuloplasty) may restore tricuspid valve com petence in patients with persistent symptoms despite treatment. In individuals with a primary leaflet pathologic condition, tricuspid valve replacement may be necessary.

Pulmonic Stenosis and Regurgitation

Multivalvular Disease Multivalvular disease is common, especially in patients with rheumatic heart disease and in the older adult population.

Section III—Cardiovascular Disease

Often, regurgitant lesions, such as TR and pulmonic regurgitation, are the result of another valve lesion, such as MS in association with pulmonary hypertension. In general, symptoms are most often related to the most proximal valve lesion. However, the severity of each individual lesion may be difficult to assess clinically, and, therefore, careful evaluation with echocardiography and right and left ventricular heart catheterization is necessary to assess valve function before any planned surgery. Failure to correct all significant valvular lesions may result in a poor clinical outcome. Double valve replacement is associated with a higher operative and long-term mortality than single valve replacement.

Rheumatic Heart Disease Acute rheumatic fever (ARF) is the sequelae of group A β-hemolytic streptococcal infection. The disease is thought to be secondary to an abnormal immunologic response to the streptococcal infection. ARF usually occurs in children 4 to 9 years of age, with boys and girls being equally affected. Although the prevalence of this disease has significantly decreased in the United States over the past several decades, it still poses a major health care problem in many developing nations, and endemic outbreaks have been identified even in the United States. ARF is characterized by a diffuse inflammation of the heart (pancarditis). An exudative pericarditis is common and often results in fibrosis and obliteration of the pericardial sac. Constrictive pericarditis is rare. The myocardium is often infiltrated with lymphocytes, and areas of necrosis may occur. The characteristic histologic finding in the myocardium is the Aschoff body, which is a confluence of monocytes and macrophages surrounded by fibrous tissue. Valvulitis is characterized by verrucous lesions on the leaflet edge, which are composed of cellular infiltrates and fibrin. The mitral valve is most frequently involved, followed by the aortic valve. Involvement of the tricuspid or pulmonic valve is rare. Valvulitis can be recognized by the presence of a new insufficiency murmur. Aortic stenosis and MS do not occur for many years, when progression of the fibrosis results in restricted leaflet mobility. The presentation of ARF is usually an acute, febrile illness 2 to 4 weeks after a streptococcal pharyngitis infection. Because the diagnosis of ARF cannot be made by laboratory tests alone, guidelines based on the symptoms and a physical examination have been established (modified Jones criteria) (Table 8-3). A diagnosis of ARF can be made Table 8-3 Revised Jones Criteria Major Criteria Carditis (pleuritic chest pain, friction rub, heart failure) Polyarthritis Chorea Erythema marginatum Subcutaneous nodules Minor Criteria Fever Arthralgia Previous rheumatic fever or known rheumatic heart disease

if two major, or one major and two minor, criteria are present after a recent, documented streptococcal pharyngitis infection. Major criteria include evidence of carditis (e.g., pleuritic chest pain, friction rub, heart failure, MR), poly arthritis, chorea, erythema marginatum, and subcutaneous nodules. Minor criteria include fever, arthralgia, and a history of rheumatic fever or known rheumatic heart disease. Once the diagnosis is established, a course of therapy with penicillin is indicated to eradicate the streptococcal infection. Salicylates are effective for the treatment of fever and arthritis. Corticosteroids and immunosuppressive therapy have not been proved beneficial in the management of the carditis. Heart failure should be treated with standard therapy. Recurrent attacks of rheumatic fever are common, especially during the first 5 to 10 years after the primary illness. Rheumatic fever prophylaxis should be continued during this period, and for 10 years in patients with a high exposure rate to streptococcal infection (e.g., health care professionals, child care workers, military recruits). Patients with significant rheumatic heart disease should receive prophylaxis indefinitely, considering the high rate of recurrence in these individuals. The recommended therapy for prophylaxis is an intramuscular injection of 1.2 million units of benzathine penicillin monthly. Alternatively, oral penicillin or erythromycin may be used. Noncompliance with these agents reduces the effectiveness of this mode of therapy.

Prosthetic Heart Valves Two types of artificial heart valves are available: mechanical valves (tilting disk and bi-leaflet) and tissue valves (bioprostheses) (Fig. 8-5). The mechanical valves have a favorable

Caged ball

Caged disk

Bi-leaflet tilting disk Tilting disk Tissue Figure 8-5 Designs and flow patterns of major categories of prosthetic heart valves: caged ball, caged disk, tilting disk, bi-leaflet tilting disk, and bioprosthetic (tissue) valves. Whereas flow in mechanical valves must course along both sides of the occluder, bioprostheses have a central flow pattern. (From Schoen FJ, Titus JL, Lawrie GM: Bioengineering aspects of heart valve replacement. Ann Biomed Eng 10:97-128, 1982; Schoen FJ: Pathology of cardiac valve replacement. In Morse D, Steiner RM, Fernandez J [eds]: Guide to Prosthetic Cardiac Valves. New York, Springer-Verlag, 1985, p 208.Copyright © 1985 Springer-Verlag.)

Chapter 8—Acquired Valvular Heart Disease hemodynamic profile and are extremely durable. However, mechanical valves carry a high thromboembolic risk and require long-term anticoagulation. Bioprosthetic use is less likely to be complicated by thromboembolic disease, but the durability of the valve is significantly less than with mechanical valves, especially in young patients. The type of prosthesis used in a particular patient is dependent on multiple factors, including the patient’s age, suitability for long-term anticoagulation, and valve position. The American College of Cardiology/American Heart Association guidelines are available for anticoagulation therapy in patients with prosthetic heart valves. All patients with mechanical valves require warfarin therapy. Aspirin (75 to 100mg/day) is recommended as sole long-term therapy in patients with biologic prosthetic valves and in combination with warfarin in patients with mechanical valves. Warfarin is recommended during the first 3 months after surgery in patients receiving biologic prostheses. The INR should be maintained at 2.3 to 3 in low-risk patients with mechanical aortic valves. High-risk features (e.g., atrial fibrillation, left ventricular dysfunction, previous thromboembolism, and hypercoagulable state) should lead to an INR goal of 2.5 to 3.5 in patients with mechanical AVR. The INR should be maintained at 2.5 to 3.5 in all patients with mechanical mitral valves. Replacement of a diseased valve with an artificial valve results in a new set of potential risks and complications with the prosthesis. All valve prostheses result in some degree of stenosis because the effective valve orifice is smaller than that of the native valve. Thrombosis or calcification of the prosthetic valve can result in prosthetic dysfunction and hemodynamically important stenosis. Prosthetic valve insufficiency can result from perivalvular leaks in the area of the sewing ring. With bioprosthetic valves, deterioration of the prosthetic valve leaflets can lead to valve insufficiency and stenosis. Hemolysis is a frequent complication of the older mechanical valves (e.g., ball cage, disk cage) and can occur with present-day prostheses if turbulent flow asso ciated with prosthetic valve dysfunction exists, especially regurgitation. Endocarditis remains a potential complication in all patients with prosthetic valves. The guidelines for endocarditis prophylaxis are provided later (see “Endocarditis Prophylaxis”). Evaluation of prosthetic valve function is best performed with two-dimensional and Doppler echocardiographic techniques. Transesophageal echocardiography is particularly useful in studying prosthetic valves when thrombosis or endocarditis is suggested. Mechanical valves can be assessed with fluoroscopy to determine whether leaflet excursion is normal. Gated cardiac computed tomography may also be helpful (Fig 8-6; Web Fig 8-1).

Endocarditis Prophylaxis Patients with valvular heart disease and prosthetic heart valves are at increased risk for developing endocarditis (Table 8-4) (see Chapter 100). In the past, endocarditis prophylaxis was widely recommended for abnormal valves. However, recently updated guidelines reflect the fact that only an extremely small number of cases of infective endocarditis may be prevented by antibiotic prophylaxis, even if it were 100% effective. Infective endocarditis is more likely

C Figure 8-6 Images of mechanical heart valves obtained with gated cardiac CT. Maximum intensity projection of bileaflet tilting disk mechanical aortic prosthesis in closed (A) and open (B) positions. Dynamic images of this valve can be seen on Web Figure 8-1. (C) Volume-rendered image showing closed, single-leaflet aortic valve and open bi-leaflet tilting disk mitral valve in diastole. Both valves were clearly seen to open and close normally on the dynamic line images. This patient had suspected prosthetic valve dysfunction that was incompletely evaluated by both fluoroscopy and transesophageal echocardiography.

Section III—Cardiovascular Disease

Table 8-4 Cardiac Conditions in which Antibiotic Prophylaxis Is Recommended Prosthetic heart valves Previous bacterial endocarditis Congenital cardiac malformations (class IIa) Unrepaired cyanotic Repaired with residual shunt First 6 months after repair with prosthetic material Cardiac transplant recipients with valve regurgitation due to structurally abnormal valve

Table 8-5 Dental and Surgical Procedures in which Endocarditis Prophylaxis Is Recommended For patients with the underlying cardiac conditions shown in Table 8-4, prophylaxis is reasonable for all dental procedures that involve manipulation of either gingival tissue or the periapical region of teeth or perforation of oral mucosa. Prophylaxis is not recommended solely on the basis of an increased lifetime risk for acquisition of infective endocarditis. Administration of antibiotics solely to prevent endocarditis is not recommended for patients who undergo a genitourinary or gastrointestinal tract procedure.

to result from frequent exposure to random bacteremia associated with daily activities than from bacteremia caused by a dental, gastrointestinal tract, or genitourinary procedure. Lastly, the risk for antibiotic-associated adverse events exceeds the benefit (if any) of prophylactic therapy. The role of antibiotic prophylaxis is to prevent infection of the abnormal valve during procedures that are associated with transient bacteremia (Table 8-5). The flora commonly found in the part of the body being instrumented determines the choice of antibiotics. All patients with known valve disease or prosthetic heart valves should carry a card indicating the nature of their valve lesion and the type of endocarditis prophylaxis recommended. The Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease of the American Heart Association does not recommend routine antibiotic prophylaxis in patients with valvular heart disease undergoing uncomplicated vaginal delivery or caesarean section unless infection is suspected. Antibiotics are optional for high-risk patients with prosthetic heart valves, a previous history of endocarditis, complex congenital heart disease, or a surgically constructed systemic-pulmonary conduit.

Prospectus for the Future Infectious disease such as ARF as the principal cause of acquired valvular heart disease will continue to decline worldwide. For isolated mitral valvular stenosis, for example, percutaneous approaches by a skilled operator will become the standard and the preferred treatment modality over a surgical procedure. Aortic and mitral regurgitation caused by degenerative and

References American College of Cardiology/American Heart Association Task Force on Practice Guidelines: 2008 Focused update incorporated into the ACC/AHA 2006 guidelines for the management of patients with valvular heart disease. J Am Coll Cardiol 32:el-e142, 2008. Freed LA, Levy D, Levine RA, et al: Prevalence and clinical outcome of mitral-valve prolapse. N Engl J Med 341:1-7, 1999. Enriquez-Sarano M, Avierinos JF, Messika-Zeitoun D, et al: Quantitative determinants of the outcome of asymptomatic mitral regurgitation. N Engl J Med 352:875-883, 2005.

other acquired diseases will create the major morbidity and mortality rates with advancing age. In the next decade, rapid advances in percutaneous approaches to address mitral regurgitation and aortic stenosis may radically change our practice and strategies for treating these conditions.

Freeman RV, Otto CM: Spectrum of calcific aortic valve disease: Pathogenesis, disease progression, and treatment strategies. Circulation 111:3316-3326, 2005. Zoghbi WA, Enriquez-Sarano M, Foster E, et al: Recommendations for evaluation of the severity of native valvular regurgitation with two-dimensional and Doppler echocardiography. J Am Soc Echocardiogr 16:777-802, 2003. ACC/AHA guidelines update on valvular heart disease: focused update on infective endocarditis. J Am Coll Cardiol 52:676-685, 2008.

Chapter

III

Coronary Heart Disease Andrew D. Michaels

Epidemiology Coronary heart disease (CHD) is the leading cause of death in the industrialized world. Apart from its influence on mortality, it causes substantial morbidity, disability, and loss of productivity. With improvements in diagnosis, prevention, and treatment, the mortality rate from CHD has declined gradually over the past several decades. Nonetheless, 1.2 million people have a myocardial infarction (MI) or fatal cardiac event each year in the United States alone. Nearly half of all deaths in industrialized nations and 25% of those in developing countries are due to CHD. By the year 2020, CHD is predicted to surpass infectious disease as the world’s leading cause of death and disability.

Pathophysiology of Atherosclerosis In the industrialized world, atherosclerosis often begins in the early decades of life. One in six American teenagers dying accidentally has pathologic evidence of coronary atherosclerosis. Several processes contribute to the initiation and progression of atherosclerosis, including accumulation of lipoproteins, endothelial injury, and inflammation. In the early phase of atherosclerosis, small lipoprotein particles penetrate the vascular endothelium, where they are oxidized and coalesce into aggregates in the intimal layer. This process is accelerated at sites of endothelial injury, which may be caused or accelerated by systemic hypertension, hypercholesterolemia, cigarette smoking, or excessive sheer forces. The accumulation of intimal lipid aggregates stimulates the expression of adhesion molecules (e.g., intracellular adhesion molecule-1, vascular cell adhesion molecule-1, selectins) on the luminal surface of the endothelial cells, thereby enabling them to bind circulating monocytes (e.g., macrophages). The adherent monocytes intercalate between the endothelial cells into the intimal layer in response to chemokines and cytokines produced by endothelial and medial smooth muscle cells. The intimal monocytes

ingest the lipoprotein aggregates to become lipid-filled monocytes, or foam cells. Aggregates of these foam cells make up the earliest visible evidence of atherosclerosis, or the fatty streak. Foam cells replicate and release proinflammatory mediators, thereby perpetuating the local inflammatory process with resultant lesion progression. In addition, they release enzymes that cause endothelial denudation. Because the endothelium is involved in the control of vascular tone through its production of vasodilating substances such as prostacyclin and nitric oxide (e.g., endothelium-derived relaxing factor) and thrombosis, injury to these cells impairs vasodilation and creates a local prothrombotic state. Circulating platelets adhere to sites of endothelial injury and release growth factors, which stimulate the migration and proliferation of smooth muscle cells and fibroblasts from the media. This leads to formation of a fibrous cap over the lipid-rich core (Web video 1—Coronary Atherosclerosis, http://www.heartsite.com/html/cad.html). As lipids continue to accumulate in the foam cells, they undergo necrosis and leave a remnant lipid pool in the core of the plaque. Metalloproteinase enzymes (e.g., collagenase, gelatinase) released by macrophages and mast cells in the plaque degrade collagen and extracellular matrix proteins adjacent to the lipid pool, whereas cytokines (e.g., interferon-α) released by T lymphocytes inhibit the formation of collagen by vascular smooth muscle cells. This combination of increased collagen degradation and decreased collagen production creates a vulnerable plaque, which is predisposed to fissure or rupture. Such vulnerable plaques have a lipid-laden core and a thin, weakened fibrous cap. When the thin fibrous cap fissures or ruptures, highly thrombogenic collagen and lipid are exposed to circulating blood with resultant adhesion of platelets and formation of an intraluminal thrombus. Activated platelets release substances (e.g., thromboxane, serotonin) that promote vasoconstriction and thrombus propagation. When the extent of platelet aggregation and thrombosis is sufficient to impair blood flow (partially or completely), an acute coronary event (unstable angina, non–ST-segment elevation myocardial infarction [NSTEMI] or ST-segment elevation myocardial infarction [STEMI]) occurs. 95

Section III—Cardiovascular Disease

Figure 9-1 Angiograms of the right coronary artery. A, Discrete stenosis is observed in the middle segment of the artery (arrow). B, Same artery is shown after successful balloon angioplasty of the stenosis and placement of an intracoronary stent.

When the atherosclerotic plaque is covered with a thick fibrous cap, rupture is less likely, but the plaque may gradually increase in size. As plaque volume increases, the coronary arterial lumen is compromised, and blood flow is impaired (Fig. 9-1). The hemodynamic significance of plaque is determined by the length and severity of the luminal narrowing; in general, a 70% decrease in the luminal diameter of a coronary artery limits blood flow in the presence of increased myocardial oxygen demands (e.g., exercise, emotional excitement), leading to the clinical condition of exertional angina. A 90% decrease in luminal diameter is sufficient to limit flow even when myocardial oxygen demands are normal. For intermediate-severity lesions with a diameter stenosis between 40% and 70%, assessing the hemodynamic significance of the stenosis with either fractional flow reserve (FFR; ratio of the mean coronary pressure distal to the stenosis [acquired by a micromanometer pressure transducer on a coronary angioplasty guidewire] divided by the mean arterial pressure proximal to the stenosis) or stress testing may help determine the requirement for coronary revascularization.

Table 9-1 Risk Factors and Markers for Coronary Artery Disease Nonmodifiable Risk Factors Age Male sex Family history of premature coronary artery disease Modifiable Independent Risk Factors Hyperlipidemia Hypertension Diabetes mellitus Metabolic syndrome Cigarette smoking Obesity Sedentary lifestyle Heavy alcohol intake Markers Elevated lipoprotein(a) Hyperhomocysteinemia Elevated high-sensitivity C-reactive protein (hsCRP) Coronary arterial calcification detected by electron-beam computed tomography (EBCT) or multidetector computed tomography (MDCT)

Risk Factors Several risk factors for the development of atherosclerosis have been identified (Table 9-1). Nonmodifiable risk factors include (1) advanced age, (2) male sex, and (3) family history of premature atherosclerosis. The prevalence of coronary artery disease (CAD) increases with age. At any given age, the prevalence is higher in men than in women. On average, the clinical manifestations of CAD become evident about 10 years later in women than in men. A family history of premature atherosclerosis (occurring in men before age 55 years and in women before age 65 years) increases the risk for atherosclerosis in an individual, likely as a result of environ-

mental factors (e.g., dietary habits, cigarette smoking) and a genetic predisposition to the disease. Other risk factors are modifiable, and their treatment may decrease the risk for atherosclerosis. These modifiable risk factors include hyperlipidemia, hypertension, diabetes mellitus, metabolic syndrome, cigarette smoking, obesity, sedentary lifestyle, and excessive alcohol intake. Although several definitions of the metabolic syndrome have been endorsed, the definition adopted by the National Cholesterol Education Program Adult Treatment Panel requires at least three of the following five criteria: waist circumference

Chapter 9—Coronary Heart Disease more than 102cm in men and more than 88cm in women; triglyceride level 150mg/dL or higher; high-density lipoprotein (HDL) cholesterol level lower than 40 mg/dL in men and lower than 50 mg/dL in women; blood pressure 130/85mmHg or higher; and serum glucose 110mg/dL or higher. Using these criteria, nearly 25% of the U.S. population has metabolic syndrome. Finally, markers associated with an increased incidence of CAD include lipoprotein(a), hyperhomocysteinemia, high-sensitivity C-reactive protein (hsCRP), and coronary arterial calcification. Lipids play a central role in the atherosclerotic process, and elevated levels of cholesterol, primarily low-density lipoprotein (LDL) cholesterol, are associated with accelerated atherosclerosis (Web video 1—Coronary Atherosclerosis, http://www.heartsite.com/html/cad.html). HDL cholesterol, by contrast, functions as a protective agent, and its serum level is inversely related to the risk of CAD. Elevated triglycerides are often associated with reduced levels of HDL cholesterol and are an independent risk factor for CAD. Large trials of lipid-lowering therapy have demonstrated the effectiveness of cholesterol reduction in the primary and secondary prevention of CAD. Systemic hypertension, defined as a systolic arterial pressure greater than 140mmHg or a diastolic pressure greater than 90mmHg, increases the risk for CAD. The risk increases proportionally with the extent of blood pressure elevation, and proper treatment of hypertension reduces the risk. Diabetes mellitus increases both the risk for developing CAD and the mortality associated with it. Although CAD is the leading cause of death in adult patients with diabetes, tight glycemic control has not been shown to reduce the risk. Diabetes mellitus often co-exists with other risk factors, including dyslipidemia (elevated triglyceride level, low HDL level), hypertension, and obesity. This grouping of risk factors has been termed the metabolic syndrome, and its presence identifies a person at increased risk for having or developing atherosclerotic disease. Cigarette smoking has adverse effects on the lipid profile, clotting factors, and platelet function and is associated with a twofold to threefold increase in the risk for CAD. Cessation of smoking reduces the excess risk for a coronary event by 50% within the first 1 to 2 years of quitting. Obesity, defined as a body mass index greater than 30 kg/m2, is often associated with other risk factors (e.g., hypertension, dyslipidemia, glucose intolerance); in addition, obesity appears to be an independent risk factor for CAD. The distribution of body fat is important, with abdominal adiposity posing a substantially greater risk for CAD in both men and women. Multiple observational studies have demonstrated an inverse relationship between the amount of physical activity and the risk for CAD. Although the ideal duration, frequency, and intensity of such physical activity have not been determined, numerous studies have shown that exercise is beneficial in healthy patients and those with or at risk for CAD. Moderate alcohol intake (1 to 2 drinks daily) is associated with a reduction in the risk for cardiovascular events; in contrast, heavy alcohol intake increases cardiovascular mortality. Lipoprotein(a) consists of LDL cholesterol linked to an apo(a) molecule. It has a homologic structure with plas-

minogen and interferes with the generation of plasmin, thereby creating a predisposition to thrombosis. Elevated levels of homocysteine are associated with an increased risk for coronary, cerebral, and peripheral vascular disease. Hyperhomocysteinemia can be treated effectively with dietary folate supplementation. However, such treatments have not been shown to reduce the incidence of stroke or cardiovascular events in patients with elevated serum homocysteine levels. CRP, a marker of inflammation, may indicate or contri bute to an increased propensity for plaque rupture and thrombosis. Elevated serum CRP levels—when measured with the new high-sensitivity assays (i.e., hsCRP)—strongly correlate with the risks for MI, stroke, peripheral arterial disease, and sudden cardiac death. Levels of hsCRP lower than 1 mg/L are associated with a low risk for vascular events; levels of 1 to 3 mg/L pose an intermediate risk; levels greater than 3 mg/L create a high risk. In healthy patients without hyperlipidemia but with an hsCRP greater than 2 mg/L, statin therapy has been shown to reduce the risk for myocardial infarction, stroke, revascularization for unstable angina, and death from cardiovas cular causes. Coronary arterial calcification is a prominent feature of coronary atherosclerosis, and it correlates with the presence and severity of CAD. Electron-beam computed tomography (EBCT) or multidetector computed tomography (MDCT) can accurately quantify coronary calcification, thereby serving as screening tests for CAD in asymptomatic patients. Currently, the usefulness of EBCT or MDCT and hsCRP in the clinical setting is poorly defined. However, the finding of coronary calcification in patients without known CAD or risk factors for CAD may identify those who warrant aggressive risk factor modification.

Nonatherosclerotic Causes of Cardiac Ischemia Although atherosclerosis is the most common disease affecting the coronary arteries, several nonatherosclerotic processes may produce myocardial ischemia or MI. Embolization from infective endocarditis, mural thrombi in the left atrium or ventricle, prosthetic valves, intracardiac tumors, or paradoxical emboli from the venous system across an atrial or a ventricular septal defect or pulmonary arteriovenous malformation may compromise coronary blood flow, leading to myocardial ischemia or MI. Chest wall trauma may result in coronary arterial dissection or thrombosis. Aortic dissection can propagate to the aortic root and occlude a coronary artery at its origin. Coronary arterial dissection may occur spontaneously during pregnancy or with connective tissue disorders such as Marfan syndrome or Ehlers-Danlos syndrome. Several forms of arteritis may involve the coronary arteries, including syphilis, Takayasu arteritis, polyarteritis nodosa, systemic lupus erythematosus, and giant cell arteritis. These syndromes may result in obstruction, occlusion, or thrombosis of the coronary arteries. Kawasaki disease, a

Section III—Cardiovascular Disease

mucocutaneous lymph node syndrome, is a systemic disease of children that causes coronary vasculitis with resultant coronary aneurysms. Spontaneous in situ coronary thrombosis may occur in the setting of hematologic disorders (e.g., polycythemia vera, essential thrombocytosis, disseminated intravascular coagulation, sickle cell anemia, paroxysmal nocturnal hemoglobinuria). Congenital coronary anomalies may cause myocardial ischemia. Spontaneous coronary spasm (e.g., Prinzmetal vasospastic angina) with or without underlying CAD may cause myocardial ischemia or, rarely, MI. Cocaine use may result in myocardial ischemia or MI through several mechanisms, including coronary vasospasm, thrombosis, and accelerated atherosclerosis. An occasional patient treated with sumatriptan for migraine headaches or paclitaxel for cancer may experience MI in the absence of CAD. In 10% to 20% of patients with suggested angina, coronary angiography reveals normal epicardial coronary arteries. In some of these individuals, microvascular or small vessel disease, the syndrome X, has been implicated. Studies have suggested that women have a higher prevalence of syndrome X compared with men presenting with possible acute coronary syndrome. The small resistance vessels in these patients, which are not visualized angiographically, have reduced vasodilatory capability. This dysfunction may lead to myocardial ischemia, as evidenced by exercise-related abnormalities on echocardiographic or nuclear scintigraphic studies. Some patients respond to treatment with common anti-anginal medications; although, in general, these drugs are less effective in patients with syndrome X than in those with atherosclerotic CAD. Finally, myocardial ischemia may result when significant increases in the demand for myocardial oxygen exceed oxygen supply. Such an oxygen supply-demand imbalance may occur in individuals with thyrotoxicosis, aortic stenosis, aortic insufficiency, tachyarrhythmia, or sepsis. Diminished oxygen supply may occur as a result of acute blood loss, hypotension, severe anemia, or carbon monoxide poisoning.

Endothelial dysfunction secondary to atherosclerosis impairs the ability of the coronary arterioles to dilate when oxygen demands increase. In addition, when a flow-limiting stenosis is present in an epicardial coronary artery, the arterioles distal to the stenosis may already be maximally or nearly maximally dilated in the resting state. The inability of the arterioles to dilate and increase coronary arterial flow during periods of increased demand (e.g., decreased coronary vasodilator reserve) results in a supply-demand mismatch, with resultant ischemia and the clinical pattern of stable angina. When myocardial oxygen supply cannot meet oxygen demand, myocardial ischemia occurs. This ischemia, in turn, initiates a series of pathophysiologic events. Regional myocardial hypoxia causes anaerobic glycolysis, lactate production, intracellular acidosis, and disordered calcium homeostasis. These intracellular changes induce abnormalities in myocardial relaxation, leading to reduced compliance and contraction, which cause regional wall motion abnormalities. Finally, electrocardiographic (ECG) evidence of ischemia (i.e., ST-segment depression or elevation) occurs, and angina pectoris ensues. If myocardial ischemia is transient, the duration of the resultant mechanical dysfunction may be short. In contrast, more prolonged ischemia may produce myocardial stunning, hibernation, or even an MI. Myocardial stunning refers to a prolonged period (e.g., hours, days) of reversible myocardial dysfunction after an ischemic event. Hibernation occurs in the setting of chronic ischemia when oxygen delivery is adequate to maintain myocardial viability but inadequate to maintain normal function. The clinical importance of the hibernating state is that restoration of blood flow to the involved myocardium results in improved mechanical function. Because of limited energy expenditure, conduction tissue is more resistant to ischemia than contractile tissue. Nevertheless, ischemia may result in altered ionic transport, altered autonomic tone, and injury to the conduction system, resulting in a variety of ischemia-induced arrhythmias and conduction abnormalities.

Pathophysiology and Consequences of Myocardial Ischemia

Angina Pectoris

In the normal myocardium, a balance between myocardial oxygen supply and demand is present at rest and during physical exertion or emotional excitement. In response to an increase in oxygen demand, an appropriate increase in oxygen supply maintains adequate tissue oxygenation. When oxygen demand increases in the setting of limited oxygen supply, myocardial ischemia results. At rest, the myocardium extracts most of the oxygen that is delivered to it through the coronary arteries. As a result, any increase in myocardial oxygen demand, as a result of an increase in heart rate, wall stress, or contractility, must be accompanied by a concomitant proportional increase in myocardial blood flow. Regulation of coronary blood flow occurs at the level of the arterioles and is dependent on autonomic tone and an intact, functioning endothelium.

For many years, patients with chronic, stable angina pectoris were believed to develop myocardial ischemia because of a transient increase in myocardial oxygen demand as a result of physical exertion or emotional excitement in the setting of limited oxygen supply caused by fixed atherosclerotic CAD. Effort-induced angina was thought to be a problem of excessive oxygen demand with limited oxygen supply. However, some patients with chronic, stable angina may develop myocardial ischemia because of dynamic coronary vasoconstriction in the setting of fixed atherosclerotic CAD. Such inappropriate coronary vasoconstriction has been shown to occur during exposure to cold, while under mental stress, and during isometric or isotonic exercise as well as during exposure to cigarette smoking. In short, chronic, stable angina is a syndrome of both increased myocardial oxygen demands in the setting of limited supply and dynamic reductions in myocardial oxygen supply, most of which are induced by common, everyday events.

Chapter 9—Coronary Heart Disease

Table 9-2 Angina Pectoris Type

Pattern

ECG

Abnormality

Medical Therapy

Stable

Stable pattern, induced by physical exertion, exposure to cold, eating, emotional stress Lasts 5-10min Relieved by rest or nitroglycerin

Baseline often normal or nonspecific ST-T changes

≥70% Luminal narrowing of one or more coronary arteries from atherosclerosis

Aspirin Sublingual nitroglycerin

Unstable

Prinzmetal or variant angina

Increase in anginal frequency, severity, or duration Angina of new onset or now occurring at low level of activity or at rest May be less responsive to sublingual nitroglycerin Angina without provocation, typically occurring at rest

Signs of previous MI ST-segment depression during angina Same as stable angina, although changes during discomfort may be more pronounced Occasional ST-segment elevation during discomfort

Plaque rupture with platelet and fibrin thrombus, causing worsening coronary obstruction

Anti-ischemic medications Statin Aspirin and clopidogrel Anti-ischemic medications Heparin or LMWH Glycoprotein IIb/IIIa inhibitors Statin

Transient ST-segment elevation during pain Often with associated AV block or ventricular arrhythmias

Coronary artery spasm

Calcium channel blockers Nitrates

AV, atrioventricular; ECG, electrocardiography; LMWH, low-molecular-weight heparin; MI, myocardial infarction.

The patient with exertional angina pectoris (Table 9-2) usually complains of a retrosternal pressure or dull ache during physical exertion, while eating, during exposure to cold, or with emotional excitement. Other adjectives that the patient may use to describe the chest discomfort include “viselike,” “constricting,” “crushing,” “heavy,” and “squeezing.” In many patients, the retrosternal pain radiates to the jaw, neck, and left shoulder and arm. Dyspnea often accompanies exertional angina pectoris and may be associated with diaphoresis and nausea. Although its duration varies considerably from one patient to another, the episode usually lasts 3 to 10 minutes. On occasion, however, it may linger for as long as 20 to 30 minutes. It is typically relieved by sublingual nitroglycerin within 1 to 3 minutes. At a time when the patient is not experiencing angina, the physical examination is usually normal. During an episode of chest discomfort, the patient may become somewhat pale and diaphoretic, and the respiratory rate and effort may increase. The heart rate and systemic arterial pressure are usually greater than at rest. Pulmonary congestion (e.g., rales at both bases posteriorly) may be evident. On auscultation of the heart, an S4 is usually audible as a result of decreased left ventricular compliance, and a transient S3 may be present if left ventricular systolic dysfunction occurs. In an occasional patient, ischemia-induced papillary muscle dysfunction will cause a murmur of mitral regurgitation to be audible at the cardiac apex. As the episode of angina resolves, the pulmonary rales, diastolic heart sounds, and systolic murmur may quickly disappear. Three noninvasive techniques have been used to demonstrate transient episodes of myocardial ischemia in the patient with exertional angina pectoris: (1) During exerciseinduced or spontaneous chest pain, the ECG usually shows ST-segment depression that is reflective of subendocardial

ischemia, which resolves within minutes of the pain’s disappearance (Fig. 9-2). (2) During episodes of angina, global left ventricular systolic function may decline, and segmental wall motion abnormalities may develop. These abnormalities can be observed with two-dimensional echocardiography, magnetic resonance imaging, or gated blood pool scintigraphy. The assessment of regional abnormalities and systolic function using two-dimensional echocardiography performed during exercise or intravenous dobutamine infusion is a particularly useful technique for detecting myocardial ischemia. As with the ECG alterations, these segmental wall motion abnormalities may resolve within minutes after relief of pain, or they may linger for hours. (3) Myocardial perfusion may be assessed during exercise-induced angina by the intravenous injection of a radioactive tracer, such as thallium-201 or technetium sestamibi, followed by imaging with the appropriate equipment.

EVALUATION OF THE PATIENT WITH ANGINA For the patient in whom the cause of chest pain is unclear, stress testing may help clarify the diagnosis by reproducing the patient’s symptoms and demonstrating objective evidence of ischemia. Submitting the patient to exercise or pharmacologic stress provides an opportunity to assess the evidence of ischemia through the evaluation of ECG abnormalities (e.g., routine stress testing), perfusion defects (e.g., radionuclide imaging), or segmental wall motion abnormalities (e.g., echocardiography). As with all diagnostic tests, the predictive value of exercise testing is influenced by the pre-test probability that the patient has CAD. For example, in the patient with a high pre-test probability of having CAD, a positive test is highly predictive, whereas a test with

100

Section III—Cardiovascular Disease

Boston University Hospital

1 MAR 1999

aVR

aVL

III

aVF

aVR

aVL

III

aVF

B Figure 9-2 Electrocardiogram obtained during angina (A) and after the administration of sublingual nitroglycerin and subsequent resolution of angina (B). During angina, transient ST-segment depression and T-wave abnormalities are present.

negative results has a high likelihood of being falsely negative. Conversely, in the individual with a low likelihood of having CAD, a negative test is highly predictive, but a positive test result has a high likelihood of being falsely positive. Stress testing may also be useful in the patient with chronic stable angina for the determination of exercise capacity, documentation of the effectiveness of medications, and risk stratification (i.e., identifying patients at risk for CAD in whom more aggressive therapies may be warranted) (Fig. 9-3). The Clinical Outcomes Utilizing Revascularization and

Aggressive Drug Evaluation (COURAGE) trial has demonstrated that patients with a large degree of ischemic myocardium during stress testing are most likely to benefit with coronary revascularization. In a patient with a normal resting ECG, routine stress testing with ECG monitoring is usually sufficient. However, in patients with baseline ECG abnormalities (e.g., nonspecific ST-segment abnormalities, left ventricular hypertrophy, left bundle branch block [LBBB], or ventricular pre-excitation) and in patients taking digoxin, the specificity of exercise-induced ST-T-wave changes is diminished. In these

Chapter 9—Coronary Heart Disease 1 min 30 sec of recovery

Standing at rest 138/90

101

166/94 HR: 98

HR: 58 Moderate chest pain 162/94

Submaximal exercise

6 min of recovery

HR: 115

HR: 82

Mild chest pain 164/94

108/90

Chest pain resolved

Maximal exercise

HR: 127

10 min of recovery 104/90 HR: 87

Moderate chest pain Figure 9-3 Treadmill exercise test demonstrates a markedly ischemic electrocardiogram (ECG) response. The resting ECG is normal. The test was stopped when the patient developed angina at a relatively low workload, accompanied by ST-segment depression in lead II and ST-segment elevation in lead V2. These changes worsened early in recovery and resolved after administration of sublingual nitroglycerin. Only leads II and V2 are shown; however, ischemic changes were seen in 10 of the 12 recorded leads. Severe atherosclerotic disease of all three coronary arteries was documented at subsequent cardiac catheterization.

individuals, echocardiographic, nuclear scintigraphic, or magnetic resonance imaging improves both the sensitivity and the specificity of stress testing, albeit at substantially increased cost. Nuclear imaging is preferred over echocardiography in patients with LBBB. Exercise-induced ECG changes in women are less specific than in men; for this reason, many physicians perform exercise testing with imaging in all women. Several prognostic markers associated with a poor clinical outcome have been identified in the patient undergoing routine stress testing; these include (1) ischemic ECG changes (ST-segment depression) that occur early in exercise, in multiple leads, or persist for several minutes after the completion of exercise; (2) an associated decrease (rather than the normal increase) in blood pressure levels; and (3) poor exercise tolerance (e.g., less than 6 minutes exercise duration on the standard Bruce protocol). In patients whose baseline ECG is sufficiently abnormal to preclude an adequate interpretation during exercise, the standard exercise test may be combined with radionuclide perfusion imaging, ECG assessment of left ventricular global and segmental function, or magnetic resonance imaging of left ventricular function. When stress testing is combined with imaging, the sensitivity for detecting CAD is about 90%, which is greater than that achieved with standard ECGguided exercise testing. The specificity is about 80%, and the predictive value is about 90%. When a radionuclide stress perfusion imaging study is performed, a radioactive tracer, such as thallium-201, technetium-99m sestamibi, or technetium-99 tetrofosmin, is immediately administered intravenously before exercise is terminated. Because the radioactive tracer is distributed to the myocardium in proportion to coronary arterial blood flow, segments of myocardium that become ischemic during exercise have decreased uptake of the radioactive tracer relative to normally perfused areas of myocardium. Within 4

hours of the injection of thallium, about 50% is redistributed throughout viable myocardium, which results in a filling in of areas that were hypoperfused at peak exercise. Unlike thallium, technetium sestamibi and tetrofosmin do not redistribute to areas that were ischemic. The presence and extent of exercise-induced perfusion abnormalities provide prognostic information. Patients with a normal stress perfusion study—with or without CAD—have an extremely low risk for future cardiac events (75th percentile of patient’s age and sex for coronary calcification), hsCRP > 3mg/dL, or metabolic syndrome. CAD, coronary artery disease; CRP, C-reactive protein; HDL, highdensity lipoprotein; hsCRP, high-sensitivity C-reactive protein; LDL, low-density lipoprotein; TG, triglycerides.

been established (Table 9-4). Patients should be instructed on dietary changes; an evaluation by a nutritionist may be helpful. Patients with CAD generally should be treated aggressively for hyperlipidemia management. Statin medications are most commonly prescribed for a goal LDL lower than 100 mg/dL. Recent studies suggest that further LDL reductions below 70mg/dL provide further risk reduction. For those CAD patients with normal cholesterol levels, statin therapy may be helpful in stabilizing atherosclerotic plaque, resulting in a reduced risk for MI and stroke. For those with low HDL, niacin or fibrate agents may be helpful, in addition to aerobic exercise, to achieve a goal HDL of greater than 40mg/dL. All patients with known or suggested CAD should be placed on antiplatelet therapy (e.g., aspirin, 75 to 325 mg daily; clopidogrel, 75mg daily for patients allergic to aspirin) unless a contraindication to antiplatelet therapy is present. These agents decrease the rates of MI and death in patients with angina or previous MI. In addition, they may decrease the risk for MI in individuals without suggested CAD but with multiple risk factors. Angiotensin-converting enzyme (ACE) inhibitors should be prescribed to patients with CAD who have diabetes mellitus or left ventricular systolic dysfunction unless contraindicated. Angiotensin receptor

Chapter 9—Coronary Heart Disease blockers may be used in patients who develop an ACE inhibitor–induced dry cough. Although exercise is often limited by angina, regular activity at a level that is tolerated should be encouraged. Isometric exercise, such as weight lifting and high-intensity activities, especially in the cold (e.g., skiing, shoveling snow), are not advisable. However, many patients with stable angina may perform vigorous activities, including moderate physical exertion at work. For obese patients and those with a sedentary lifestyle, regular aerobic activity is recommended. As previously noted, the pathophysiologic characteristics of angina are one of supply-demand mismatch. Therefore, its therapy is directed at correcting the mismatch by decreasing myocardial oxygen demands, augmenting myocardial oxygen supply, or both. Nitrates, β blockers, and calcium channel blockers are among the pharmacologic options most commonly used for the control of symptoms in patients with chronic stable angina (Table 9-5). They appear to be of similar efficacy in controlling anginal symptoms. When a single agent fails to control angina, combination therapy is usually effective. Ranolazine, a selective inhibitor of late sodium influx, is an effective antianginal agent that has no effect on heart rate or systemic blood pressure. Ranolazine may be used as either a first- or second-line agent for patients with angina. Unlike aspirin and lipid-lowering therapy, none of these agents has been convincingly shown to decrease mortality in patients with CAD.

103

Nitrate preparations have been used in the medical management of exertional angina for many years. The effect of nitrates is mediated through relaxation of vascular smooth muscle. Dilation of arterioles reduces systemic vascular resistance and therefore afterload. Nitrates have a more pronounced effect on the venous system; venodilation results in venous pooling, decreased venous return, and therefore decreased preload. The arteriolar and venodilatory effects substantially reduce myocardial oxygen demands, thereby decreasing angina. In addition, nitrates augment coronary blood flow by dilating epicardial coronary arteries (although this effect is minimal in extensively diseased arteries) and increasing blood flow through collateral vessels. Several formulations are available. Sublingual nitroglycerin tablets or oral spray is effective for the acute treatment of anginal episodes and as prophylactic therapy before an activity that is likely to provoke angina. Topical nitroglycerin ointment and oral preparations are effective for the chronic management of stable angina, whereas intravenous nitroglycerin is appropriate for patients with unstable angina and acute MI. The chronic use of nitrates may result in tolerance, an effect that can be minimized by allowing for a daily nitrate-free period; for example, removing topical nitrate preparations during sleeping hours or prescribing oral nitrates that avoid aroundthe-clock administration. β-adrenergic blocking drugs are competitive inhibitors of catecholamine β receptors. They decrease myocardial oxygen

Table 9-5 Medications for Angina Pectoris Drug Class

Examples

Nitroglycerin

Sublingual Topical Intravenous Oral

β-adrenergic blocking agents

Metoprolol Atenolol Propranolol

Calcium channel blocking agents

Nadolol Phenylalkylamine (verapamil) Benzothiazepine (diltiazem)

Calcium channel blocking agents

Dihydropyridine (nifedipine, amlodipine)

Late sodium current blocking agents

Ranolazine

AV, atrioventricular; INa, sodium current.

Anti-anginal Effect

Physiologic Side Effects

Decreased preload and afterload Coronary vasodilation Increased collateral blood flow Decreased heart rate Decreased blood pressure Decreased contractility

Headache Flushing Orthostasis

Tolerance develops with continuous use

Bradycardia Hypotension

May worsen heart failure and AV conduction block; avoid in vasospastic angina

Decreased heart rate Decreased blood pressure Decreased contractility Coronary vasodilation Decreased blood pressure Coronary vasodilation Inhibits cardiac late INa Prevents calcium overload

Bronchospasm Depression Bradycardia Hypotension

Comments

May worsen heart failure and AV conduction

Constipation with verapamil

Hypotension, reflex tachycardia Peripheral edema Dizziness Headache Constipation Nausea

Short-acting nifedipine associated with increased risk for cardiovascular events No effects on blood pressure or heart rate Modest QTc prolongation

104

Section III—Cardiovascular Disease

demands by reducing heart rate, blood pressure, and contractility. These agents are effective in controlling anginal symptoms (especially exercise-induced symptoms), and they decrease mortality and reinfarction in survivors of MI. β blockers differ in their lipid solubility, duration of action, and β-receptor selectivity. β1 receptors predominate in the heart, where they mediate increases in heart rate, contractility, and atrioventricular (AV) conduction. β2 receptors predominate in the vascular and bronchial smooth muscle. Blockade of β1 receptors produces several beneficial cardiac effects, whereas β2-receptor blockade may induce broncho spasm and peripheral vasoconstriction. Atenolol and metoprolol are β1 selective at low doses; however, at the moderate to high doses often used in clinical practice, all β blockers lose their selectivity. Because β blockers may worsen underlying conduction system abnormalities, they should be used with caution in patients with conduction system dysfunction. In addition, these agents may result in a mild increase in the triglyceride level and a mild decrease in the HDL cholesterol level. Calcium ions play a critical role in myocardial and vascular smooth muscle contraction and in the genesis of the cardiac action potential (see Chapter 10). Blocking these effects with a calcium antagonist results in a decrease in heart rate, myocardial contractility, and peripheral arterial vasodilation, all of which decrease myocardial oxygen demand. In addition, coronary vasodilation occurs, resulting in augmented oxygen supply. Three major classes of calcium antagonists are available, and the specific agent of choice should be individualized for the particular patient. The dihydropyridine medications (e.g., nifedipine, amlodipine) predominantly cause vasodilation with little or no effect on heart rate, contractility, or AV conduction. In fact, the vasodilation may lead to a reflex tachycardia. The phenyl alkylamine medications (e.g., verapamil) reduce heart rate, slow AV conduction, depress contractility, and have less of an effect on peripheral vascular tone than the dihydropyridine medications. They may not be tolerated in patients with depressed ventricular systolic function or underlying conduction system disease. The benzothiazepine medications (e.g., diltiazem) have less vasodilatory action than the dihydropyridine medications and less myocardial suppressant action than the phenylalkylamine medications.

REVASCULARIZATION IN PATIENTS WITH ANGINA In patients for whom medical therapy does not effectively control anginal symptoms and in patients considered to be at increased risk clinically (e.g., unstable angina, angina associated with heart failure, angina associated with arrhythmia, poor exercise capacity) or by noninvasive testing (e.g., large amount of ischemic myocardium, depressed left ventricular systolic function), coronary revascularization plays an important therapeutic role. Several modalities for coronary revascularization exist, including surgical revascularization (e.g., coronary artery bypass grafting [CABG]) and catheter-based percutaneous techniques (e.g., percutaneous coronary intervention [PCI]). With advances in equipment, adjunctive pharmacologic agents, and increasing operator experience, PCI can now be achieved with high success rates and at relatively low risk.

More than 1 million PCI procedures are performed each year in the United States alone. (Web video 9-1—Angioplasty, http://www.heartsite.com/html/ptca.html). With percutaneous transluminal coronary angioplasty (PTCA), a highpressure inflation of a distensible balloon is performed at the site of coronary arterial narrowing with resultant enlargement of the lumen. Balloon inflation causes denudation of the endothelial surface, fracture of the atherosclerotic plaque, and disruption of the vessel intima. The vessel lumen can be successfully dilated in greater than 90% of cases. In 2% to 5% of patients undergoing PTCA, the coronary arterial injury is severe, and, as a result, the artery occludes abruptly. Such patients are usually treated with intracoronary stenting or rarely urgent CABG to prevent acute MI. In patients in whom PTCA is initially successful, up to 50% develop restenosis at the site of balloon dilation within 1 to 6 months of the angioplasty. Of the patients who develop restenosis, about 50% experience recurrent angina. Restenosis is a complex process involving elastic recoil of the artery, vascular remodeling, and hyperplasia of the vascular intima. It is not prevented by the administration of antiplatelet, anticoagulant, anti-anginal, or hypolipidemic medications. During the past decade, intracoronary stenting now has become the technique of choice for PCI (Web video 9-2—Intracoronary Stenting, http://www.heartsite.com/ html/stent.html). The coronary stent—a cylindric, expandable metal structure available in varying diameters and lengths—is premounted on an angioplasty balloon. When the balloon is positioned at the site of the stenosis and inflated to expand the stent, the stent becomes permanently embedded in the arterial wall. Subsequently, the balloon is deflated and removed, but the stent maintains its expanded cylindric configuration, thereby acting as a scaffold to maintain vessel patency. In this way, stenting results in a greater increase in luminal size than can be achieved with balloon angioplasty alone (see Fig. 9-1B). Stents can be used to treat PTCA-related coronary arterial dissections, thereby avoiding the need for urgent CABG. In comparison with balloon angioplasty, stenting is associated with a reduced incidence of abrupt closure (about 1% to 2%) and restenosis (about 20% to 25%), thereby explaining why it is the procedure of choice in more than 90% of PCIs. At the same time, stenting may not be the procedure of choice in small coronary arteries (luminal diameter < 2.0 mm) because these vessels are too small for the smallest available stents. The person in whom intracoronary bare metal stenting (BMS) has been performed should receive aspirin indefinitely and clopi dogrel for 2 to 4 weeks to prevent thrombosis. During the weeks after stent deployment, the stent becomes endothelialized, at which time it is no longer thrombogenic or subject to abrupt closure. Studies have demonstrated that a 1-year course of clopidogrel is superior compared with a 4-week course after bare metal stenting, with a reduction in adverse ischemic events. Beginning in 2003, drug-eluting stents (DES) have been coated with antiproliferative drugs (e.g., sirolimus [Rapamycin], paclitaxel [Taxol]), which are extremely effective in preventing restenosis. There are five FDA-approved DES in the United States, listed in order of their approval with the year of approval: Cordis Cypher (sirolimus; 2003), Boston Scientific Taxus (paclitaxel; 2004), Medtronic Endeavor (zotarolimus; 2007), Abbott Xience V (everolimus;

Chapter 9—Coronary Heart Disease 2008), and Boston Scientific Promus (everolimus; 2008). The incidence of restenosis with DES is 5% to 10%. Ran domized trials have demonstrated a roughly 70% reduction in stent restenosis with DES compared with BMS. In stenting procedures in the United States, roughly 65% involve DES, and 35% involve BMS. The person who receives a DES should take aspirin indefinitely and clopidogrel for at least 12 months. The antiproliferative agent that coats the stent delays the process of endothelialization; as a result, these stents are subject to thrombosis and abrupt closure for months after their placement. Other percutaneous interventional techniques that have a limited role in coronary revascularization include rotational and directional atherectomy; thrombectomy; brachytherapy, which is the application of local radiation therapy to treat restenosis after stenting; and coronary laser therapy. Of these, thrombectomy has the most promising role in PCI procedures for STEMI patients. Studies performed in the 1970s and 1980s established the effectiveness of CABG for the control of anginal symptoms and, in some patients, offered an improvement in survival when compared with anti-anginal medical therapy. Harvesting a segment of saphenous vein or radial artery and anastomosing it to the ascending aorta (proximally) and the distal portion of the diseased coronary artery (distally) is performed with CABG. The internal mammary artery can be dissected free from the pleural surface and its distal end anastomosed to a diseased coronary artery. These procedures effectively bypass the sites of atherosclerotic narrowing, thereby allowing blood to flow freely to the myocardium perfused by the diseased artery. The left internal mammary artery is most commonly used to bypass the left anterior descending coronary artery, and has 10-year patency rates of roughly 90%. In comparison, the patency rate for saphenous vein grafts is 50% at 10 years. Whenever possible, the mammary artery is used because its long-term patency is superior to that of venous or radial arterial conduits. Experienced surgeons perform CABG with a peri-operative mortality rate of 1% to 2%, a stroke rate of 1% to 2%, and a peri-operative MI rate of 5% to 10%. CABG improves survival (when compared with medical therapy) in patients with greater than 50% luminal diameter narrowing of the left main coronary artery or narrowing of all three major epicardial coronary arteries in conjunction with mildly or moderately depressed left ventricular systolic function (e.g., ejection fraction, 35% to 50%). In addition, CABG improves long-term survival in patients with a narrowing of two or three epicardial coronary arteries and normal left ventricular systolic performance, provided that the proximal portion of the left anterior descending coronary artery is significantly narrowed. In the short term (within 1 to 2 years of the procedure), those having PCI are more likely than those undergoing CABG to require anti-anginal medications or a subsequent revascularization procedure largely because of the incidence of symptomatic restenosis after successful percutaneous coronary revascularization. Because of the progressive decline of graft patency between 5 and 10 years postoperatively, the benefits of surgery over percutaneous revascularization are less apparent in the long term. Small, randomized studies comparing the two approaches to revascularization in patients with multivessel CAD and preserved left ven-

105

tricular systolic function (e.g., ejection fraction > 50%) demonstrate no difference in mortality after 1 to 5 years of follow-up, except in patients with diabetes, who fare better with CABG. Recently, however, larger observational studies showed that CABG is associated with higher long-term survival than stenting in patients with multivessel CAD. A recently completed trial showed no difference between CABG and DES for left main or multivessel CAD in mor tality or myocardial infarction. DES did have higher rates of repeat revascularization because of restenosis, but lower rates of stroke. There are ongoing randomized trials in diabetic patients with multivessel CAD randomized to CABG or DES. Unfortunately, neither percutaneous nor surgical revascularization techniques halt the underlying atherosclerotic process, and new stenoses may develop at previously uninvolved sites in native coronary arteries and in the bypass grafts. Aspirin should be administered immediately after CABG and continued thereafter because it improves graft patency. If a stenosis develops in a bypass graft, percutaneous revascularization is often effective. In addition, repeat CABG is possible, although the surgical risks are higher than with the first procedure. For patients with severe angina refractory to maximal medical therapy and coronary revascularization, treatment options include external counterpulsation and spinal cord stimulation. External counterpulsation involves inflation of three lower extremity cuffs during diastole and deflation during systole. This treatment is performed in 1-hour sessions for a total of 35 treatments. Roughly 75% of patients report an improvement in angina severity, and the treatment is generally well-tolerated. The likely mechanism of action involves improved endothelial function. Spinal cord stimulation provides analgesia for patients with severe angina by placing a stimulating electrode in the dorsal epidural space at the C7-T1 level. Although preliminary data appear promising, there is a paucity of data on intermediate- or longterm outcomes. Transmyocardial laser revascularization is no longer recommended for refractory angina.

VARIANT ANGINA In 1959, Prinzmetal and colleagues described a group of patients with variant angina. These patients usually experienced chest pain at rest rather than with physical exertion or emotional excitement, and the ECG recorded during chest pain showed ST-segment elevation rather than depression, which resolved as the pain subsided (Fig. 9-4). On occasion, episodes of chest discomfort were accompanied by varying degrees of AV block or ventricular ectopy, but MI was uncommon. Patients with variant angina did not often have the usual risk factors for coronary atherosclerosis, although cigarette smoking was frequent. Subsequent angiographic studies demonstrated that variant angina is the result of epicardial coronary arterial spasm, which may occur either at the site of an atherosclerotic plaque or in the setting of angiographically normal coronary arteries. During coronary angiography, coronary vasospasm may be provoked by the intracoronary infusion of acetylcholine or ergonovine. In addition, methacholine, a parasympathomimetic agent, has been used to induce coronary arterial spasm, similar to the arterial spasm in response to exposure

106

Section III—Cardiovascular Disease

Figure 9-4 Continuous electrocardiogram recording in a patient with Prinzmetal (variant) angina. The spontaneous onset of chest discomfort began during the top strip, accompanied by transient ST-segment elevation. By the bottom strip, several minutes later, both discomfort and ST-segment elevation had resolved.

to cold (e.g., cold pressor test), the production of a significant alkalosis (e.g., vigorous hyperventilation during the intravenous infusion of an alkalotic buffer solution), and histamine administration. Calcium channel blockers, alone or in combination with long-acting nitrate preparations, are highly effective in patients with coronary arterial spasm. They are the treatment of choice for patients with variant angina. β blockers are contraindicated in patients with vasospastic angina because blockade of the vasodilatory effects of β2-receptor stimulation may result in unopposed α-adrenergic vasoconstriction. For the rare patient who has continued episodes of coronary arterial spasm despite maximal medical therapy, intracoronary stenting may be performed.

Acute Coronary Syndromes The term acute coronary syndrome encompasses the clinical syndromes of unstable angina, NSTEMI, and STEMI. Patients with unstable angina or NSTEMI are usually indistinguishable by history, physical examination, and ECG findings. The distinction between these two groups is made only after the results of the serum cardiac enzyme analyses are available. The patient with unstable angina or NSTEMI may develop myocardial ischemia or MI through several mechanisms. Most commonly, these individuals have subendocardial ischemia or necrosis as a result of decreased coronary blood flow, which is due to platelet aggregation or a partially occlusive intracoronary thrombus at the site of an ulcerated atherosclerotic plaque. In addition, concomitant plateletmediated coronary arterial vasoconstriction at the site of plaque ulceration may occur. Alternatively, the patient may develop myocardial ischemia or MI because of an increase in myocardial oxygen demand that cannot be met by an appropriate increase in coronary blood flow. In some indi-

viduals, the coronary blood flow cannot appropriately increase because of severe CAD. In those without CAD, subendocardial ischemia or infarction may occur solely as a result of significantly augmented myocardial oxygen demands in the setting of a normal supply (e.g., uncontrolled hypertension, thyrotoxicosis) or a decrease in myocardial oxygen delivery (e.g., profound anemia, hypoxemia). Patients with left ventricular hypertrophy (due to hypertension or aortic stenosis) are at increased risk for subendocardial ischemia. The patient with unstable angina pectoris usually complains of retrosternal chest pain similar in character and consistency to that of the patient with stable, exertional chest pain. In contrast to the patient whose angina is stable, however, these individuals usually report that their anginal frequency, severity, or duration has worsened, and they may report pain at rest. Furthermore, the patient may note that nitroglycerin is ineffective or less effective in relieving the chest pain. On physical examination, the patient may exhibit no visible or audible abnormalities at a time when he or she is pain free. During an episode of chest pain, however, the patient may become anxious, diaphoretic, and dyspneic. The heart rate often increases, although bradycardia may occur secondary to enhanced vagal tone or transient AV block and most commonly with inferior wall ischemia or infarction. On cardiac auscultation, an S4 may be audible as a result of decreased left ventricular compliance. An S3 may be present if left ventricular systolic dysfunction occurs, and a systolic murmur of mitral valve papillary muscle dysfunction may be appreciated. Evidence of pulmonary congestion is often present and may reflect an elevated left ventricular filling pressure as a result of decreased left ventricular compliance or systolic dysfunction. If a large area of myocardium is involved and left ventricular systolic dysfunction ensues, then frank pulmonary edema may occur. For the patient who is experiencing chest pain, a 12-lead ECG should be immediately obtained because it is frequently diagnostic of myocardial ischemia or MI and is important

Chapter 9—Coronary Heart Disease 10 mm/mV 25 mm/s Filter ON 1-11-111

aVR aVL aVF

10 mm/mV V1-V2-V3

107

V4-V5-V6

Figure 9-5 Acute anterolateral myocardial infarction. Leads I, aVL, and V2 to V6 demonstrate ST-segment elevation. Reciprocal ST-segment depression is seen in leads II, III, and aVF. Deep Q waves have developed in leads V2 and V3.

in determining the appropriate treatment plan. STEMI, previously referred to by the pathologically inaccurate term, transmural infarction or Q-wave myocardial infarction, refers to an acute coronary syndrome in which ST-segment elevation (e.g., ST-segment elevation at the J point in two contiguous leads at least 0.2mV in men or at least 0.15mV in women in leads V2 to V3 or at least 0.1mV in other leads) is present on the surface ECG. These infarctions typically are the result of complete thrombotic occlusion of a coronary artery and may first be exhibited on the ECG by symmetrically peaked or hyperacute T waves. These peaked T waves resolve after several minutes as the characteristic ST-segment elevation develops (Fig. 9-5). The distribution of leads with ST-segment elevation can identify the myo cardial location and the culprit coronary artery: anterior MI, V2 to V5, left anterior descending coronary artery; inferior MI, II, III, aVF, right coronary artery; lateral MI, I, aVL, V6, left circumflex or diagonal; posterior, V7 to V9, left circumflex coronary artery; right ventricular RV4, right coronary artery. NSTEMI, previously termed subendocardial infarction or non–Q-wave myocardial infarction, and unstable angina occur as a result of a subtotally occlusive thrombus or a thrombus that was initially totally occlusive but not sustained, enabling partial or complete lysis to occur within minutes to hours of its formation. They are associated with ST-segment depression and T-wave inversions on the surface ECG (Fig. 9-6). In one fourth to one half of patients with acute MI, the first ECG does not demonstrate typical ST-segment changes. In this situation, serial ECGs should be obtained to increase the diagnostic yield. If a patient has ongoing chest pain without ST-segment changes, posterior lead ECG recording of leads V7 to V9 should be performed. A posterior lead ECG is used to assess for posterior wall injury, usually the result of left circumflex coronary artery occlusion, which is not readily apparent on a standard 12-lead ECG. If acute MI is

suggestive but the initial ECG does not confirm the diagnosis, demonstration of new regional wall motion abnormalities with echocardiography may be helpful in confirming the diagnosis. Myocardial necrosis results in the release of certain intracellular enzymes into the blood. Their appearance in the blood allows the identification of myocardial necrosis, and their quantitation over a number of hours allows for the estimate of its amount. Because 20% of patients with acute MI have atypical or no symptoms (i.e., silent MI) and the initial ECG is nondiagnostic in up to 50% of patients, serologic identification of myocyte necrosis has become an important diagnostic tool. Several serum markers have been identified (Fig. 9-7). Creatine kinase (CK) and its myocardial-specific isoenzyme, creatine kinase muscle band (CK-MB) are detectable in the blood within 3 to 6 hours of the onset of MI. They reach their peak concentration at 24 hours and return to normal within 48 hours. Although CK-MB is relatively specific for cardiac injury, it may be elevated in subjects with extensive skeletal muscle injury or disease, chronic renal disease, or hypothyroidism. Troponins I and T are regulatory proteins involved in the interaction of cardiac actin and myosin. Because they are not present to any extent in other organs and are not detectable in blood under normal circumstances, an increase in their serum concentration is more specific and sensitive for myocyte necrosis than an increase in the concentration of other enzymes. After cardiac injury, the serum troponin concentration begins to rise within 4 to 6 hours and remains elevated for 7 to 10 days. False-positive elevations of troponin T, but not troponin I, have been observed in patients with renal failure. The presence of heterophilic antibodies or fibrin may interfere with the assay for troponin I and give false-positive results. The former is found in 3% of the general population and a high percentage of patients with autoimmune disease; the latter may be found in blood that

108

Section III—Cardiovascular Disease

Boston University Hospital I

aVR

aVL

III

aVF

Figure 9-6 Marked ST-segment depression in a patient with prolonged chest pain is the result of an acute non–ST-segment elevation myocardial infarction (NSTEMI). Between 1 and 3mm of ST-segment depression is seen in leads I, aVL, and V4 to V6. The patient is known to have had a previous inferior myocardial infarction.

heart rate or blood pressure, pulmonary embolism, hypoxemia) or reduced supply (i.e., hypotension) in the absence of epicardial CAD.

CPK

Serum enzyme level

TREATMENT OF UNSTABLE ANGINA AND NON–ST-SEGMENT ELEVATION MYOCARDIAL INFARCTION cTnl, cTnT

LDH AST Normal

10 15 Days after infarction Figure 9-7 Typical time course for the detection of enzymes released after myocardial infarction. AST, serum aspartate aminotransferase; CPK, creatine kinase; cTnI, cardiac troponin I; cTnT, cardiac troponin T; LDH, lactate dehydrogenase.

has been heparinized. Because serum troponin concentration is an extremely sensitive measure of myocardial necrosis, such elevations are sometimes observed in patients with myocardial necrosis as a result of increased myocardial oxygen demand (i.e., subjects with a significantly elevated

Unstable angina and NSTEMI may be clinically indistinguishable with ECG studies. They are differentiated only by the presence of serologic evidence of myocardial necrosis. Accordingly, the initial treatment of these patients is similar and includes (1) hospital admission with serial assessment of ECGs and sequential measurements of cardiac enzymes; (2) aggressive anti-anginal, antiplatelet, and antithrombotic therapy; and (3) identification of the patient at increased risk for recurrent ischemia, MI, or death who may benefit from revascularization. With optimal medical therapy, the 1-year mortality rate of patients with unstable angina or NSTEMI is 3% to 5% (Fig. 9-8). Rest for 24 to 48 hours with continuous ECG monitoring, analgesics, and supplemental oxygen therapy are frequently prescribed. Sublingual nitroglycerin should be given initially, followed by oral or topical nitroglycerin. Intravenous nitroglycerin should be administered if recurrent chest pain occurs. In the absence of a contraindication, β blockers should be promptly instituted because they decrease heart rate and blood pressure levels and left ventricular contractility, thereby reducing myocardial oxygen demand. The calcium antagonists, verapamil and diltiazem, may be useful for the patient who fails to respond to nitrates and β blockers as well as for the patient with a contraindication to β blockers. However, calcium antagonists should not be used in the patient with known depressed left ventricular systolic function or with evidence of pulmonary vascular congestion on

Chapter 9—Coronary Heart Disease

109

Symptoms suggestive of an acute coronary syndrome

1. 12-Lead electrocardiogram 2. Aspirin 3. Supplemental oxygen 4. Sublingual nitroglycerin 5. Morphine PRN 6. Cardiac enzymes

No ST-segment changes Initial enzymes normal

Observe 1. Nitroglycerin PRN 2. Analgesia 3. Serial ECG and cardiac enzymes

ST-segment elevation consistent with STEMI

1. Clopidogrel 2. LMWH

1. IV nitroglycerin 2. Heparin 3. IV β blocker

Recurrent chest pain or serial studies positive

No high-risk markers No recurrent chest pain

No recurrent chest pain, serial studies negative

Provocative stress testing Negative results

ST-segment depression consistent with unstable angina or NSTEMI

Provocative stress testing Negative results Positive results

Positive results

High-risk markers present - elevated troponin - persistent/recurrent chest pain - persistent ST depression - associated heart failure - hemodynamic instability - LVEF 4 hours’ duration). t-PA and its derivatives do not elicit an antibody response or hypotension. t-PA is given as an initial bolus or front loaded, followed by a 90-minute infusion (15mg as a bolus, another 50mg infused over 30 minutes, and the remaining 35mg infused over the next 60 minutes). r-PA is administered as a double bolus (two 10-unit boluses delivered 30 minutes apart), and TNK-tPA is administered as a single bolus (0.5mg/kg to a maximum of 50 mg). Although r-PA and TNK-tPA are somewhat more likely than t-PA to restore early patency of the infarct-related artery, the mortality rate with these three agents is similar. Overall, thrombolytic therapy decreases short-term mortality in subjects with STEMI by about 20%. Angiographic studies comparing thrombolytic regimens demonstrate that restoration of blood flow in the infarct-related artery is faster and more complete with t-PA than with streptokinase, and this translates into a modestly decreased mortality rate with t-PA, especially when it is given in a front-loaded fashion. Specifically, in the Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries (GUSTO) trial, t-PA was associated with a statistically significant 1% absolute reduction in mortality when compared with streptokinase. Most of this benefit occurred in patients younger than 70 years within 4 hours of the onset of an anterior STEMI. In older patients, in patients more than 4 hours after symptom onset, and in those with an MI in a territory other than the anterior wall, the mortality difference between these two agents was negligible. The contraindications to thrombolytic therapy are listed in Table 9-8; they identify those with an unacceptably high risk for bleeding complications. The most catastrophic potential complication of thrombolytic therapy is intra cranial hemorrhage. This risk is substantially increased in patients with a history of hemorrhagic stroke, uncontrolled hypertension, body weight less than 70 kg, and age more than 65 years. Aspirin is an obligatory adjunct to thrombolysis; its use is associated with an additive benefit on mortality and a decrease in recurrent ischemic events. Clopidogrel therapy

Chapter 9—Coronary Heart Disease Table 9-8 Contraindications to Thrombolytic Therapy in Acute Myocardial Infarction

Table 9-9 Complications of Acute Myocardial Infarction

Absolute

Left ventricular failure Right ventricular failure Cardiogenic shock

Suspected aortic dissection Active bleeding* Any prior cerebral hemorrhage Intracranial neoplasm Cerebral aneurysm or arteriovenous malformation Ischemic cerebrovascular accident within 3 months Relative Bleeding diathesis, coagulopathy, or anticoagulant use Major surgery within 3 weeks Puncture of a noncompressible vessel, internal bleeding, or head or major body trauma within previous 2 weeks Nonhemorrhagic stroke or gastrointestinal hemorrhage within 6 months Proliferative retinopathy Active peptic ulcer disease History of chronic, severe, poorly controlled hypertension Severe uncontrolled hypertension on presentation (systolic blood pressure > 180mmHg or diastolic blood pressure >110mmHg) Traumatic or prolonged (>10min) cardiopulmonary resuscitation Pregnancy *Does not include menstrual bleeding.

with a 300-mg load followed by a daily maintenance dose of 75mg improves mortality compared with placebo in STEMI patients 75 years and older. Intravenous heparin administered for 48 hours is necessary to maintain patency of the infarct-related artery after successful thrombolysis when a t-PA is administered, but not with streptokinase. Lowmolecular-weight heparin may be slightly more effective than unfractionated heparin as adjunctive therapy after successful thrombolysis. Its use is associated with a higher rate of vessel patency and a lower rate of reocclusion, leading to fewer episodes of recurrent ischemia and infarction, albeit with a somewhat increased risk for hemorrhagic complications. For patients who do not reperfuse within 90 minutes of receiving thrombolytics (as evidenced by continued chest pain or continued ST-segment elevation), rescue PCI is generally recommended. Even with successful thrombolysis, there is a 30% chance of culprit vessel reocclusion within 3 months. Risk stratification with a submaximal, symptomlimited stress test or coronary angiography is required during the STEMI hospitalization for thrombolized patients.

POST–MYOCARDIAL INFARCTION COMPLICATIONS The complications of MI may be categorized as electrical or mechanical (Table 9-9).

Arrhythmias and Conduction Abnormalities Cardiac arrhythmias may occur in patients with acute coronary syndromes. Those that cause symptoms or hemodynamic compromise almost always warrant treatment, whereas those that do not often can be managed expectantly.

113

Functional

Mechanical Free-wall rupture Ventricular septal defect Papillary muscle rupture with acute mitral regurgitation Electrical Bradyarrhythmias (first-, second-, and third-degree atrioventricular blocks) Tachyarrhythmias (supraventricular, ventricular) Conduction abnormalities (bundle branch and fascicular blocks)

Although most of these arrhythmias are a direct result of the ischemic process, other reversible aggravating factors, such as electrolyte disturbances, hypoxemia, and medication toxicity, must be excluded. Premature ventricular complexes, ventricular couplets, and nonsustained ventricular tachycardia (VT) occur frequently in the peri-infarction period. Although such ectopy can be effectively suppressed with anti-arrhythmic agents, treatment is not warranted in the absence of symptoms or hemodynamic compromise. The presence of frequent ventricular ectopy does not predict the development of more malignant arrhythmias, and empiric therapy of such ectopy is associated with an increased mortality rate. Accelerated idioventricular rhythm, or slow VT, often occurs shortly after successful reperfusion, is self-limited, and does not require treatment. During the past several decades, mortality in hospitalized patients with acute MI has substantially declined in large part because of the early recognition and treatment of lethal arrhythmias. Because most deaths from acute MI occur as a result of sustained VT or ventricular fibrillation (VF), these rhythm disturbances should be treated with immediate electrical defibrillation, after which administering intravenous anti-arrhythmic medications (e.g., lidocaine, amio darone) is reasonable for 24 to 48 hours. Sustained but hemodynamically stable VT can be treated initially with anti-arrhythmic agents with electrical cardioversion held in reserve. In the absence of electrolyte abnormalities, polymorphic VT is usually a marker of recurrent or persistent ischemia, and aggressive anti-ischemic treatment is warranted. When sustained VT or VF occurs in the first 48 hours after MI, it does not portend the same poor prognosis as it does when it occurs later. Transient supraventricular tachyarrhythmias may occur in patients with acute MI, with sinus tachycardia and atrial fibrillation being the most common. The cause of sinus tachycardia (e.g., anxiety, pain, fever, anemia, hypoxemia, hypovolemia, pulmonary vascular congestion, thyrotoxicosis) should be promptly identified and corrected. If atrial fibrillation is accompanied by a rapid ventricular response, with resultant ongoing ischemia or hemodynamic compromise, electrical shock cardioversion should be considered. In the patient with atrial fibrillation and a rapid ventricular

114

Section III—Cardiovascular Disease

response, intravenous β blockers or amiodarone is usually effective for controlling the ventricular response, provided no contraindications to their use exist. Calcium channel blocking agents are also effective but should be avoided in the patient with heart failure. (These arrhythmias are discussed at length in Chapter 10.) Bradyarrhythmias may complicate acute MI. The most common bradyarrhythmia is sinus bradycardia, which is observed in 20% to 25% of patients with acute MI and is more common in those with inferior than anterior MI. In patients with inferior MI, sinus bradycardia is often associated with hypotension caused by increased vagal tone as a result of stimulation of vagal afferent fibers in the inferoposterior portion of the left ventricle (Bezold-Jarisch reflex). Unless accompanied by hemodynamic instability, sinus bradycardia should be simply observed. If treatment is necessary, intravenous atropine (0.5 to 2mg) should be administered, aiming for a heart rate of 60 beats/minute and a resolution of symptoms. Temporary pacing is rarely required. Ischemia and infarction can result in transient or permanent injury to the conduction system. Varying degrees of AV block may occur in patients with acute MI. Ischemia of the AV node can result in first-degree or Mobitz type I second-degree (Wenckebach phenomenon) AV block. These rhythms are most often associated with inferior MI; they are transient, do not adversely affect survival, and do not require treatment unless the ventricular rate is sufficiently slow to produce syncope, congestive heart failure, or angina. Mobitz type II second-degree AV block is a rare complication of acute MI (1% of cases) and usually results from injury to the His-Purkinje system in the setting of an extensive anterior MI. It often is associated with progression to complete or third-degree AV block and is an indication for temporary transvenous or transcutaneous pacing in anticipation of implantation of a permanent pacemaker. Complete or thirddegree AV block may occur with inferior or anterior MI. When it occurs in the setting of an inferior MI, the block is usually at the level of the AV node. It is associated with a stable escape rhythm and tends to be transient, although it may take up to 1 to 2 weeks to resolve. As a result, treatment with only a temporary and not a permanent pacemaker is usually required. In contrast, when complete AV block occurs in the setting of an anterior MI, the His-Purkinje system is usually involved. The block is usually permanent, and a permanent pacemaker should be implanted. Block in one or more branches of the conduction system may occur with acute MI and is more common with anterior than with inferior infarction. Patients with isolated left anterior or left posterior fascicular block or right bundle branch block (RBBB) do not require specific therapy. Conversely, temporary pacing is suggested in patients with new bifascicular blocks (e.g., LBBB or RBBB with left anterior or left posterior fascicular block) because progression to complete heart block is common. If bifascicular block persists after an MI, a permanent pacemaker should be placed.

Congestive Heart Failure and Cardiogenic Shock Patients who die of cardiac failure after acute MI have extensive myocardial necrosis with loss of at least 40% of the functioning left ventricular muscle mass, as either a conse-

quence of new infarction or a combination of old and new infarctions. The patient with an acute MI and no evidence on physical examination or chest radiographic studies of left ventricular failure has an excellent prognosis, with only a 2% to 5% inhospital mortality rate (Killip class I). The individual with some evidence of pulmonary vascular congestion (e.g., basilar rales, S3, radiographic evidence of pulmonary venous congestion) is classified as Killip class II and has a short-term mortality rate of 10% to 15%. In the patient with overt pulmonary edema evidenced on physical examination or chest radiographic studies, mortality rate is 20% to 30% (Killip class III). Finally, the patient with cardiogenic shock is said to be Killip class IV and has a mortality rate of 50% to 60% even with maximal therapy. In these individuals, infarction is associated with systemic arterial hypotension and diminished peripheral perfusion, as manifested by mental confusion, cold and clammy skin, peripheral cyanosis, and oliguria. Hemodynamically, the systemic arterial systolic pressure is less than 90mmHg, the cardiac index is less than 1.8L/m2 per minute, the systemic arteriolar resistance is greatly increased (>2000 dynes/cm5 per second), and the left ventricular filling pressure is elevated (more than 20 mm Hg) for more than 30 minutes. The reduced systemic arterial pressure further diminishes coronary arterial perfusion pressure, thereby increasing myocardial ischemia. The low cardiac output and systemic arterial pressure induce an intense sympathetic discharge that produces peripheral vasoconstriction, further decreasing tissue perfusion and causing a systemic lactic acidosis, which depresses myocardial function. In response to a reduced cardiac output, the heart rate increases, thereby increasing myocardial oxygen demand. As left ventricular filling pressure rises, subendocardial perfusion is further compromised. In short, the hemodynamic and metabolic consequences of cardiogenic shock cause worsening myocardial ischemic injury, which, in turn, leads to worsening left ventricular dysfunction. A cycle of severe hemodynamic impairment and deteriorating myocardial oxygenation is established. The therapy of the patient with an acute MI and resultant left ventricular dysfunction depends on the extent of such dysfunction. The normotensive individual with symptoms and signs of Killip class II congestive heart failure (i.e., mild orthopnea, basilar rales, S3) usually responds to bed rest, salt restriction, a loop diuretic, and low-dose vasodilator therapy with an ACE inhibitor. Additional therapy with digitalis or other inotropic agents is not usually necessary, nor is invasive hemodynamic monitoring. The management of the patient with more severe heart failure (Killip class III or IV) should be based on a careful assessment of hemodynamic variables obtained with a balloon-tipped flotation catheter in the pulmonary artery and an intra-arterial cannula. Adequate oxygenation should be ensured by continuous pulse oximetry with supplemental oxygen or ventilator support as needed. Placement of a urinary catheter enables the urine output to be assessed accurately, and endotracheal intubation and assisted ventilation may reduce the work of breathing and improve tissue oxygenation. Intravenous furosemide should be administered in an attempt to reduce the pulmonary capillary wedge pressure in the range of 18 to 20mmHg; this appears to be the optimal preload in the setting of acute MI. In the normotensive individual, vasodilator therapy

Chapter 9—Coronary Heart Disease with nitroglycerin and oral ACE inhibitors should be instituted to reduce afterload, increase cardiac output, and lower left ventricular filling pressure. The resultant decrease in left ventricular wall stress reduces myocardial oxygen requirements, improves subendocardial perfusion, and helps relieve ischemia. Nitroglycerin should be administered to avoid an excessive reduction in systemic arterial pressure, which may compromise myocardial perfusion, while keeping pulmonary capillary wedge pressure in the range of 18 to 20mmHg. The patient with heart failure and hypotension or an inadequate response to diuretics and vasodilators (i.e., cardiac output < 1.8 L/m2 per minute pulmonary capillary wedge pressure > 20mmHg) should be treated with intravenous inotropic agents (e.g., dopamine or dobutamine, depending on the systemic arterial pressure). If the patient is only mildly hypotensive, dobutamine is the preferred inotropic agent. Dopamine should be reserved for the patient with more severe hypotension because it may increase pulmonary capillary wedge pressure. In the patient with severe heart failure or cardiogenic shock, a careful search for a potentially correctable cause should be undertaken. Two-dimensional and color Doppler echocardiography, which can rapidly be performed at the bedside, will allow the clinician to determine whether the shock is due to extensive left ventricular dysfunction or a mechanical problem, such as acute mitral regurgitation, acute ventricular septal defect, extensive right ventricular infarction, or a contained rupture of the left ventricular free wall (see “Mechanical Complications”). Patients with shock who are examined within the first few hours of the onset of MI should be considered for immediate reperfusion therapy. Thrombolytic agents are less effective in opening the occluded infarct-related artery in the patient with cardiogenic shock, and these agents have not convincingly exerted a beneficial effect in such patients. Conversely, early coronary revascularization within 12 hours of the onset of cardiogenic shock, accomplished percutaneously or sur gically, has been shown to improve in-hospital and 1-year survival. By reducing afterload and increasing myocardial perfusion pressure, intra-aortic balloon counterpulsation may be effective in stabilizing the patient with cardiogenic shock. Although initial hemodynamic improvement in this setting may be dramatic, balloon counterpulsation alone probably does not improve the poor prognosis associated with car diogenic shock. Rather, counterpulsation should be con sidered a supportive measure in patients with potentially reversible abnormalities before cardiac catheterization, cardiac surgery, or, in some cases, cardiac transplantation. In patients who remain hemodynamically unstable despite pressors and intra-aortic balloon counterpulsation, per cutaneous left ventricular support device implantation should be considered. If these therapies do not improve hemodynamics, a surgical ventricular assist device could be considered.

Right Ventricular Infarction Right ventricular infarction usually occurs in association with inferior MI because the blood supply to both these areas usually comes from the right coronary artery. The presence of a concomitant right ventricular infarction substantially increases the mortality of an inferior MI. Right ventricular

115

MI results in the clinical triad of hypotension, clear lungs (i.e., normal pulmonary capillary wedge pressure), and prominent jugular venous distention. In the absence of hemodynamic measurements, right ventricular infarction may be confused with hypovolemia, pulmonary embolism, or cardiac tamponade. In fact, the patient with acute right ventricular failure may have a prominent y descent in the atrial pressure tracing (Fig. 9-10), Kussmaul sign, and pulsus paradoxus, all of which mimic pericardial tamponade. Demonstrating ST-segment elevation in the right precordial leads (e.g., >0.1 mV elevation in V4R) confirms the diagnosis of right ventricular infarction. For this reason, a right-sided precordial ECG should be obtained in all patients with inferior MI. The treatment of hypotension in the patient with right ventricular infarction often requires rapid intravascular volume repletion (with a goal right atrial pressure of 12 to 15 mm Hg) and inotropic agents (e.g., dobutamine). A balloon-tipped flotation catheter in the pulmonary artery and an intra-arterial cannula should be used for hemodynamic monitoring. Diuretic and vasodilator (e.g., nitroglycerin) therapy should be avoided because they may provoke hypotension in this setting. If the patient can be supported during the first few days of hemodynamic instability, considerable improvement in right ventricular function often occurs.

ECG

RADIAL ARTERIAL PRESSURE

100 50 0

PULMONARY ARTERY PRESSURE

RIGHT ATRIAL PRESSURE

20 10 0 10 5 0

Figure 9-10 Electrocardiographic (ECG), arterial, and SwanGanz bedside catheter recordings in a patient with right ventricular infarction. Hypotension is present, and cardiac output, estimated by thermodilution (not shown), is reduced. The pulmonary arterial pressures are normal, whereas the right atrial pressure is elevated, and it demonstrates a prominent y descent.

116

Section III—Cardiovascular Disease

Mechanical Complications Mechanical complications of acute MI include papillary muscle rupture, ventricular septal defect, and ventricular free-wall rupture. Patients with these complications frequently experience hemodynamic collapse 3 to 5 days after acute MI. These complications are associated with high mortality rates; they account for about 15% of the mortality from acute MI. Successful immediate reperfusion therapy has reduced the appearance of these complications. Patients with late or unsuccessful reperfusion therapy are at higher risk for these mechanical complications. Papillary muscle rupture results in acute mitral regurgitation. The resultant sudden increase in left atrial volume causes a significantly elevated left atrial pressure, with resultant pulmonary edema. Papillary muscle rupture occurs most commonly with inferior MI because the posteromedial papillary muscle usually has a single source of blood supply from the right coronary artery. Conversely, the anterolateral papillary muscle has a dual blood supply. A loud, apical holosystolic murmur is usually audible, although an occasional patient with severe mitral regurgitation has no audible murmur. The diagnosis may be rapidly confirmed with transthoracic echocardiography or right-heart ventricular catheterization, with the latter demonstrating large v waves in the pulmonary capillary wedge tracing in the absence of an oxygen step-up in the right ventricle. An acute ventricular septal defect may occur after anterior or inferior MI. On physical examination, a harsh holosys tolic murmur is audible at the left lower sternal border, which may be difficult to differentiate from acute mitral regurgitation; this murmur is often accompanied by a palpable thrill. The diagnosis can be confirmed by obtaining blood samples from each of the cardiac chambers during right heart ventricular catheterization and by demonstrating a higher oxygen saturation in the samples obtained from the right ventricle or pulmonary artery compared with those obtained from the right atrium (e.g., oxygen step-up). Specifically, an increase in oxygen saturation of more than 6% between the right atrium and pulmonary artery strongly suggests the presence of a ventricular septal defect with concomitant left-to-right shunting. Doppler echocardiography also allows visualization of left-to-right shunting of blood through the ventricular septal defect. Treatment of acute papillary muscle rupture or ventricular septal defect includes inotropic agents, vasodilators, and intra-aortic balloon counterpulsation. These temporizing measures help prepare the patient for urgent cardiac surgery to repair the ventricular septal defect or replace the mitral valve. Free-wall rupture of the left ventricle almost always results in hemopericardium, cardiac tamponade, and electromechanical dissociation. Survival is uncommon and depends on prompt recognition and emergent surgical repair. In an occasional patient, a pseudoaneurysm or false aneurysm develops when free-wall rupture occurs, so that the rupture is confined by the adherent pericardium, organized thrombus, and hematoma. Because the wall of the pseudoaneurysm contains no myocardium, it may rupture at a later date. The pseudoaneurysm maintains continuity with the left ventricular cavity through a narrow connecting orifice (e.g., neck). In contrast, a true aneurysm represents an area of infarcted myocardium that has become thinned and dilated

through a process of ventricular remodeling. True aneurysms have a wide orifice or neck, their walls always contain some myocardial elements, and they rarely rupture. Pseudoaneurysms should undergo prompt surgical resection because of the risk for rupture. Conversely, a true aneurysm does not require urgent surgical resection. Medical treatment of post-MI aneurysms includes aggressive heart failure management with ACE inhibitors, β blockers, aldosterone inhibitors for those with class III or IV heart failure, and consideration of anticoagulant therapy. Those undergoing subsequent CABG and those with aneurysm-related chest pain may be considered for aneurysmectomy.

SECONDARY PREVENTION OF ACUTE CORONARY SYNDROME Secondary prevention therapies are a critical component of the management of all patients with acute coronary syndrome. About 70% of coronary heart disease deaths and 50% of MIs occur in patients with a prior history of coronary artery disease. Secondary prevention therapies in patients recovering from acute coronary syndrome represent a major opportunity to reduce cardiovascular morbidity and mortality. Before hospital discharge, patients should be educated regarding adherence to the recommended lifestyle changes and pharmacologic therapies. Patients and their families should receive discharge instructions about recognizing acute cardiac symptoms and appropriate actions to take in order to ensure early evaluation and treatment should symptoms recur. Family members should be advised about cardiopulmonary resuscitation and automatic external defibrillator (AED) training programs. Lipid management involves dietary therapy that is low in saturated fat and cholesterol (10%). Other clinical factors that increase the risk for stroke in patients with AF include prior stroke, diabetes, hypertension, heart failure, left atrial enlargement, and increasing age. No difference in stroke rate occurs between paroxysmal and chronic AF. Restoration of normal sinus rhythm has not been shown to reduce the risk for stroke. In fact, in the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) trial, a trend toward a higher incidence of stroke in patients randomized to rhythm control was found when compared with rate control, albeit not statistically significant. This trend was most likely caused by the decreased use of warfarin in the rhythm control group. Therefore, any patient with paroxysmal, persistent, or permanent AF who does not have a contraindication to anticoagulation should be treated with warfarin therapy with a target international normalized ratio between 2 and 3. Rate Control Rate control in AF is important for several reasons. It has been shown to improve symptoms and quality of life. Symptoms and hemodynamic compromise are increased at faster ventricular rates, and the tachycardic response may induce ischemia in patients with coronary artery disease. In addition, the poorly controlled heart rate may result in the development of progressive ventricular dysfunction. The heart rate can usually be controlled with digoxin, β blockers, or calcium channel blockers. In rare instances, the ventricular rate cannot be controlled by pharmacologic means, and catheter ablation of the AV node and permanent pacemaker implantation are necessary for adequate heart rate control. Occasionally, patients exhibit AF and a relatively slow ventricular rate in the absence of rate-lowering medications. This circumstance usually reflects significant underlying conduction system disease that also often involves the sinus node. Rhythm Control Rhythm control has several advantages, including (1) abolition of symptoms, (2) halting atrial enlargement (an independent predictor of stroke), and (3) improvement of left

Chapter 10—Cardiac Arrhythmias ventricular function and exercise capacity. The main disadvantage is subjecting patients to a drug therapy or procedure that might be associated with complications. As stated before, rhythm control has not been shown to reduce the risk for stroke or to have an impact on mortality when compared with rate control. Therefore, rhythm control should be attempted in patients who are symptomatic despite rate control and in those who have left ventricular dysfunction. When AF is associated with hemodynamic compromise, electrical cardioversion (with 100 to 360 joules) is the treatment of choice. In hemodynamically stable patients with less than 48 hours of AF, the risk for thromboembolism is low, and pharmacologic or electrical cardioversion can be attempted without the need for 3 weeks of anticoagulation (see later discussion). Patients with more than 48 hours of AF, or in whom the duration of the arrhythmia is unknown, are at increased risk for atrial thrombi and should be treated with anticoagulation for at least 3 weeks before an attempt at cardioversion. An alternative approach is to perform a transesophageal echocardiogram; if atrial thrombi are not present, cardioversion can be safely performed. Anticoagulation should be continued for at least 4 weeks after successful cardioversion because effective atrial contraction may be slow to return. The options for maintaining rhythm control include (1) pharmacologic therapy, (2) catheter ablation, and (3) surgical Maze procedure. The class IA (quinidine, procainamide, and disopyramide), class IC (propafenone and flecainide), and class III (sotalol and amiodarone) agents are effective in restoring sinus rhythm and for long-term maintenance therapy. However, the benefits of such therapy must be weighed against the risks for toxicity with these agents, and the probability of maintaining sinus rhythm must be taken into account. The preferred choice of drug therapy for the maintenance of sinus rhythm in patients with paroxysmal and persistent AF, based on the most recent American College of Cardiology/American Heart Association guidelines, is provided in Figure 10-7. Radiofrequency

127

ablation of the ostia of the pulmonary veins or electrical isolation of the pulmonary veins from the left atrium is a procedure that should be reserved to symptomatic patients who have failed drug therapy. Ablation therapy frequently abolishes the arrhythmia, improving symptoms and, at times, left ventricular function in patients with baseline congestive heart failure. The surgical maze procedure involves making surgical lesions in the atria that interrupt reentrant circuits and may restore sinus rhythm in more than 90% of patients. This procedure is usually performed in conjunction with mitral valve surgery.

ATRIOVENTRICULAR NODAL (JUNCTIONAL) RHYTHM DISTURBANCES Atrioventricular Nodal Reentrant Tachycardia Atrioventricular nodal reentrant tachycardia (AVNRT) is the most common type of paroxysmal SVT and is characterized by the sudden onset and termination of a regular narrow QRS complex tachycardia at rates of 150 to 250 beats/minute (Fig. 10-8A). A wide QRS complex may occur if aberrant conduction occurs in the His-Purkinje system. These rhythms may occur at any age, are somewhat more common in women than men, may occur in the absence of organic heart disease, may be short lived or sustained, and may produce palpitations, chest pain, dyspnea, and presyncope. The substrate for this tachycardia is dual AV node pathways with different effective refractory period (ERP): a fast pathway with a longer ERP and a slow pathway with a shorter ERP. Whether these pathways are exclusively intra nodal or not remains controversial, but catheter ablation studies of these pathways have revealed distinct atrial insertion sites, with the fast pathway inserting anteriorly near the His bundle and the slow pathway posteriorly near the coronary sinus ostium.

Heart disease

Yes

No (or minimal)

Flecainide Propafenone Sotalol

Heart failure

CAD

Hypertension

Amiodarone Dofetilide

Sotalol Dofetilide

LVH ≥1.4 cm

Yes Amiodarone Dofetilide

Catheter ablation

Amiodarone

Flecainide Propafenone Sotalol Amiodarone

Figure 10-7 Drug therapy to maintain sinus rhythm in patients with recurrent paroxysmal or persistent atrial fibrillation. CAD, coronary artery disease; LVH, left ventricular hypertrophy.

Catheter ablation

128

Section III—Cardiovascular Disease

F Figure 10-8 Atrioventricular (AV) nodal (junctional) rhythm disturbances. A, AV nodal reentrant tachycardia at a rate of 185 beats/ minute. The retrograde P waves are hidden in the QRS complexes. B, Automatic junctional tachycardia. Note the presence of AV dissociation during tachycardia. The P waves (arrows) are dissociated from the QRS complexes. C, Normal sinus rhythm in a patient with Wolff-Parkinson-White (WPW) syndrome. Note the short PR interval (5cm for all others Aortic arch >5.5cm Descending thoracic aorta >5cm Abdominal aorta >5.5cm Iliac aneurysm >3cm

palpable but above the umbilicus. Hypotension and acute abdominal pain should prompt consideration of aneurysm rupture, which requires emergent operative repair. Duplex ultrasonography is an accurate and reliable diagnostic tool for abdominal aortic and iliac aneurysms. Routine screening for AAA with ultrasonography is recommended for all men between the ages of 65 and 75 years or men above the age of 60 with family history of AAA among firstdegree relatives. Such screening has a proven mortality benefit. CT and MR angiography allow visualization of the thoracic and abdominal aorta as well as the iliac arteries and its branches (Fig. 13-2). Medical treatment for aortic aneurysm includes smoking cessation, tight BP control, and cholesterol reduction. β-Adrenergic blockade reduces the rate of aortic root enlargement in patients with Marfan syndrome but has not proved beneficial in patients with AAA from other causes. Patients with large aneurysms or rapid aneurysm expansion regardless of the size should undergo aneurysm repair (Table 13-2). Elective AAA repair carries a

168

Section III—Cardiovascular Disease

peri-operative mortality rate of 2% to 6%. Furthermore, a large randomized study failed to demonstrate any benefit of surgery in patients with aneurysms 4 to 5.5cm in diameter. For these reasons, patients with small aortic aneurysms should be treated medically with close monitoring of aneurysm size with periodic imaging studies every 6 to 12 months (see Table 13-2). Percutaneous endovascular aneurysm repair (EVAR) is an alternative method to open surgical repair for treatment of AAA. EVAR offers lower perioperative death than surgical repair, but long-term survival rates are not different from surgery. EVAR has not been shown to improve mortality in patients with multiple co-morbidities, who are considered to be unfit for surgery, when compared with conservative management. Therefore, it should be offered only to selected patients with symptoms from compression of adjacent organs or vascular complications.

Aortic Dissection In aortic dissection, the intimal layer is torn from the aortic wall, leading to the formation of a false lumen in parallel with the true lumen. Risk factors include hypertension, cocaine use, trauma, hereditary connective tissue disease (e.g., Marfan syndrome, Ehlers-Danlos syndrome), vasculitis (e.g., Takayasu arteritis, giant cell arteritis), Behçet disease, bicuspid aortic valve, and aortic coarctation. Aortic dissection can be classified as types A and B (Stanford system). Type A dissection involves the ascending aorta, whereas type B dissection involves the distal aorta. The DeBakey system subdivides aortic dissection into three subtypes: types I, II, and III. Type I dissection involves the entire aorta, whereas type II involves only the ascending aorta, and type III involves only the descending aorta. Aortic dissection involving the ascending aorta carries a high mortality rate of 1% to 2% per hour during the first 24 to 48 hours. Patients usually develop acute onset of severe chest or back pain. Abdominal pain, syncope, and stroke are common. Retrograde propagation of the dissection can cause pericardial tamponade or coronary artery dissection with acute myocardial infarction. Dissection involving the aortic valve causes acute severe aortic insufficiency with acute pulmonary edema. The dissection plane may propagate in an antegrade direction to compromise flow in the carotid and subclavian arteries, producing a stroke or acute upper limb ischemia. Patients with distal (type B) aortic dissection exhibit acute onset of back pain or chest pain often accompanied by lower extremity ischemia and ischemic neuropathy. The physical findings include pulse deficits, neurologic deficits, or a diastolic murmur of aortic regurgitation. However, acute aortic regurgitation into an unprepared ventricle produces only a short, soft diastolic murmur that is often missed. The widened pulse pressure and associated physical findings of chronic aortic regurgitation are absent, and the clinical picture is that of an acutely ill patient with tachypnea, tachycardia, and a narrow pulse pressure. Hypotension, jugular venous distention, and pulsus paradoxus should prompt the diagnosis of pericardial tamponade. Transesophageal echocardiography, MR angiography, or CT angiography confirms the diagnosis by demonstrating an intimal flap that separates the true lumen from the false lumen (Fig. 13-3). Type A aortic dissection is uniformly fatal without emergent surgical repair. With surgery, mortality is reduced to 10% at

Figure 13-3 Computed tomographic angiogram of the aorta shows type B aortic dissection. The intimal flap (arrow) separates the true lumen (T) from the false lumen (F) and compromises blood flow to the right kidney, causing renal atrophy and cortical thinning. (Image courtesy of Bart Domatch, MD, Radiology Department, University of Texas Southwestern Medical Center, Dallas, Texas.)

24 hours and 20% at 30 days. Patients with type B aortic dissection should be treated medically because 1-year survival is higher with medical therapy than it is with surgery (75% versus 50%). However, surgery is indicated if type B dissection compromises blood flow to the legs, kidneys, or other viscera. Tight control of BP is essential because aortic aneurysm develops in 30% to 50% of patients with type B aortic dissection studied for 4 years.

Penetrating Aortic Ulcers and Intramural Hematoma Penetrating aortic ulcers and intramural hematomas exhibit chest pain that is indistinguishable from that of aortic dissection. In contrast to aortic dissection, however, the pathologic condition is localized. No identifiable intimal flap and thus no branch vessel occlusion are produced. Disruption of the internal elastic lamina produces aortic ulcers that erode into the medial wall and protrude into the surrounding structures. Rupture of the vasa vasorum causes formation of localized hematoma underneath the adventitia with resultant asymmetrical thickening of the aortic wall. Patients with either condition typically are older than those with aortic dissection, have a larger aortic size, and have a higher prevalence of AAA. Aortic rupture is the major complication of both penetrating ulcers and intramural hematomas, particularly with those aneurysms located in the ascending aorta. The diagnosis is made with invasive angiography, CT angiography, or MR angiography (Fig. 13-4). Surgical intervention should be considered for ulcers and hematomas of the ascending aorta, deeply penetrating ulcers, or severely bulging hematomas, irrespective of their location. Ulcers and hematomas of the descending aorta may be managed successfully with β-adrenergic blockade and tight control of BP.

Chapter 13—Vascular Diseases and Hypertension

Figure 13-4 Computed tomographic angiogram of the descending thoracic aorta shows a large penetrating aortic ulcer above the diaphragm (arrow). (Image courtesy of Bart Domatch, MD, Radiology Department, University of Texas Southwestern Medical Center, Dallas, Texas.)

Other Arterial Diseases Buerger disease is a nonatherosclerotic disease of the arteries, veins, and nerves of the arms and legs affecting mostly young men before the age of 45 years. The cause is unknown, but all patients have a history of heavy tobacco addiction. The presenting symptom is claudication of the feet, legs, hands, or arms. Multiple-limb involvement and superficial thrombophlebitis are common. The C-reactive protein and Westergren sedimentation rate typically are normal, and a search for serologic markers for connective tissue disease (e.g., antinuclear antibody or rheumatoid factor, antiphospholipid antibody) is negative. The diagnosis is usually based on the typical clinical presentation. If the presentation is atypical, biopsy may be needed to make the diagnosis. The histologic hallmark is inflammatory intramural thrombi within the arteries and veins with sparing of internal elastic lamina and other arterial wall structures. The most effective treatment of Buerger disease is complete tobacco abstinence. The prostacyclin analogue iloprost constitutes adjunctive therapy to reduce limb ischemia and improve wound healing. Raynaud phenomenon is a vasospastic disease of the small arteries of mainly the fingers and toes. Primary (idiopathic) Raynaud phenomenon occurs in the absence of underlying disorders. Secondary Raynaud phenomenon occurs in association with connective tissue diseases (e.g., scleroderma, polymyositis, rheumatoid arthritis, systemic lupus erythematosus) as well as with repeated mild physical trauma (e.g., use of jack hammers), certain drugs (e.g., antineo plastic chemotherapeutic agents, interferon, monoaminereuptake inhibitors such as tricyclic antidepressants, serotonin agonists), and Buerger disease. Patients usually complain of recurrent episodes of digital ischemia, with a characteristic white-blue-red color sequence. Pallor is followed by cyanosis if ischemia is prolonged and then by erythema (reactive hyperemia) when the episode resolves. Episodes are precipitated by cold temperature or emotional stress. Physical examination can be entirely normal between

169

attacks with normal radial, ulnar, and pedal pulses. Some patients may have digital ulcers or thickening of fat pad (sclerodactyly). Patients should be instructed to avoid cold temperatures and dress warmly. Calcium channel blockers (CCBs) reduce the frequency and severity of vasospastic episodes. Giant cell arteritis is an immune-mediated vasculitis predominantly involving medium-sized and large arteries such as the subclavian artery, axillary artery, and aorta of the older adult with a strong male predominance. About 40% of patients with giant cell arteritis also have polymyalgia rheumatica, a syndrome characterized by severe stiffness and pain originating in the muscles of the shoulders and pelvic girdle. Patients may exhibit headache from temporal arteritis, jaw claudication from ischemia of the masseter muscles, or visual loss from involvement of ophthalmic artery. Chest pain suggests the co-existence of aortic aneurysm or dissection. Physical findings include low-grade fever, scalp tenderness in the temporal area, pale and edematous fundi, or a diastolic murmur of aortic regurgitation. BP difference of more than 15 mm Hg between arms suggests subclavian artery stenosis. Laboratory findings include significantly elevated C-reactive protein and Westergren sedimentation rate plus anemia. The diagnosis is confirmed by histologic examination of the arterial tissue (frequently from temporal artery biopsy), showing infiltration of lymphocytes and macrophages (i.e., giant cells) in all layers of the vascular wall. High-dose corticosteroids are highly effective. To minimize complications from long-term corticosteroid administration, the steroid dose should be tapered to find the lowest dose needed to suppress symptoms, which often wane. Every attempt should be made to discontinue corticosteroids over time. Takayasu arteritis is an idiopathic granulomatous vasculitis of the aorta, its main branches, and the pulmonary artery. This condition is particularly common in young women of Asian descent, but it also occurs in occidental women and men. The inflammatory process in the vascular wall can lead to stenosis and aneurysm formation. Hypertension, as a result of renal artery stenosis or aortic coarctation, is the most common manifestation and is present in as many as 80% of affected individuals. Because the vascular involvement is so widespread, patients may have symptoms and signs of coronary ischemia, congestive heart failure, stroke, vertebrobasilar insufficiency, or intermittent claudication. Physical findings include bruits over the subclavian arteries or aorta as well as diminished brachial pulses and thus a low brachial artery BP. The diagnosis is based primarily on this clinical presentation. First-line treatment is with corticosteroids. Other immunosuppressive agents such as methotrexate or cyclophosphamide are often added to prevent disease progression and relapse. Immunosuppressive therapy does not cause regression of preexisting vascular stenoses or aneurysms. For this reason, percutaneous or surgical revascularization is usually required. Arteriovenous (AV) fistulas are abnormal vascular communications that shunt blood flow from the arterial system directly into the venous system, bypassing the capillary beds that normally ensure optimal tissue perfusion and nutrient exchange. AV fistulas may be congenital, as in AV malformation (AVM), or acquired. The main causes of acquired AV fistula are penetrating trauma (e.g., gunshot, knife

170

Section III—Cardiovascular Disease

wound) and surgically created shunts for hemodialysis access. Patients may exhibit a pulsatile mass, symptoms related to compression of an adjacent organ, or bleeding from spontaneous rupture of an AVM. Systolic and diastolic bruits or thrills may be detectable over the fistula or AVM. An AVM in skeletal muscle may lead to bone malformation or a pathologic fracture, whereas AVM in the brain may result in neurologic deficits or seizures. High-output heart failure is another complication from a large AVM or fistula. MR angiography, CT angiography, or conventional angiography confirms the diagnosis. Depending on the size and location of the AVM, treatment options include surgical resection, transcatheter embolization, or pulse laser irradiation. Patients with acquired AV fistulas from trauma usually need surgical closure.

Pulmonary Vascular Disease Pulmonary hypertension is characterized by elevated mean pulmonary pressure of greater than 25 mm Hg at rest or greater than 30 mm Hg during exercise. The many causes of pulmonary hypertension are summarized in Table 13-3. Patients with pulmonary hypertension have not only an elevated pulmonary arterial pressure but also a low cardiac output, causing symptoms of exertional dyspnea, fatigue, and syncope. Pulmonary capillary wedge pressure is usually

Table 13-3 Classification of Pulmonary Hypertension Category 1: Pulmonary Arterial Hypertension Primary pulmonary hypertension or idiopathic pulmonary hypertension: Sporadic Familial PPH associated with: Connective tissue disease Congenital heart disease Portal hypertension Human immunodeficiency viral infection Drugs and toxins: Anorexigens Cocaine Category 2: Pulmonary Venous Hypertension Left ventricular heart failure Left ventricular valvular heart disease Category 3: Pulmonary Hypertension associated with Chronic Respiratory Disease or Hypoxemia Chronic obstructive pulmonary disease Obstructive sleep apnea Category 4: Pulmonary Hypertension associated with Chronic Venous Thromboembolism Left ventricular valvular heart disease Category 5: Pulmonary Hypertension Due to Miscellaneous Disorders Directly Affecting the Pulmonary Vasculature Sarcoidosis Histiocytosis X Compression of pulmonary vessels (adenopathy, tumor, fibrosing mediastinitis)

normal except in patients with pulmonary venous hypertension and congenital heart disease.

PULMONARY ARTERIAL HYPERTENSION Pulmonary arterial hypertension (PAH) is caused by a combination of pulmonary vasoconstriction, endothelial cell or smooth muscle proliferation, intimal fibrosis, and thrombosis in the pulmonary capillaries and arterioles. PAH is either idiopathic (primary pulmonary hypertension [PPH]) or secondary to connective tissue disease, congenital heart disease, portal hypertension, or human immunodeficiency virus (HIV) infection as well as anorexigenic drugs or toxins. Connective tissue diseases, particularly scleroderma, are the most common secondary causes of PAH. Patients with mild PAH can be asymptomatic, but patients with more advanced disease complain of dyspnea, chest pain, syncope, or presyncope. Physical findings include a left parasternal lift, loud pulmonary component of the second heart sound, murmur of tricuspid or pulmonic regurgitation, hepatomegaly, peripheral edema, or ascites. Associated ECG abnormalities indicate right ventricular hypertrophy, right atrial enlargement, or right axis deviation. Echocardiography provides important information about the severity of the pulmonary hypertension (i.e., estimated pulmonary artery pressure, right ventricular dimensions and function) and its potential causes (e.g., left ventricular failure, valvular lesions, congenital heart disease with left-to-right shunts). ) Pulmonary function tests, ventilation-perfusion ( V Q lung scans, polysomnography or overnight oximetry, autoantibody tests, HIV serology, and liver function tests also should be performed to determine other potential causes. Right ventricular catheterization should be performed in all patients with suggested PAH. Under basal conditions in the catheterization laboratory, an elevated mean pulmonary artery pressure exceeding 25mmHg, a pulmonary capillary wedge pressure below 15mmHg, and a pulmonary vascular resistance exceeding 3 units confirm the diagnosis. Acute vasodilator drug challenge should be performed during right ventricular catheterization to guide appropriate treatment. Without treatment, the prognosis of PAH is poor, with a median survival of less than 3 years. Patients with severe symptoms should be treated with prostacyclin or epoprostenol (an intravenous prostacyclin analogue) because of their proven efficacy in improving exercise capacity, quality of life, and survival. Other prostacyclin analogues, such as treprostinil and iloprost, are also effective in reducing pulmonary artery pressure and improving exercise capacity. Other classes of drugs approved for treatment of PAH include endothelin receptor blockers and phosphodi esterase-5 inhibitors. Oral CCBs are indicated for the small subset of patients with mild to moderate symptoms who demonstrate significant reduction in pulmonary pressure with acute CCB challenge. Supplemental home oxygen is indicated for all patients with hypoxemia. Higher elevations exacerbate hypoxemia, and relocation to sea level improves symptoms. Oral anticoagulation is recommended for all patients with PAH. Diuretics should be prescribed for patients with peripheral edema or hepatic congestion. Lung transplantation is recommended only for patients in whom severe symptoms occur despite intensive medical therapy.

Chapter 13—Vascular Diseases and Hypertension

Venous Thromboembolic Disease Venous thromboembolism (VTE) encompasses both deep vein thrombosis (DVT) and pulmonary embolism (PE). Among the adult U.S. population, the overall combined annual incidence is as high as 1 new case per 1000 persons. The incidence of VTE is higher in men than in women and higher in African Americans and whites than in Asians and Hispanics. More than 150 years ago, Dr. Rudolf Virchow recognized three predisposing factors: (1) endothelial damage, (2) venous stasis, and (3) hypercoagulation (Virchow triad). Endothelial damage is common with surgery or trauma, venous stasis is common with prolonged bed rest or immobilization (leg cast), and hypercoagulation is common with cancer. Trousseau syndrome consists of migratory thrombophlebitis with noninfectious vegetations on the heart valves (marantic endocarditis) typically in the setting of mucin-secreting adenocarcinoma. Dr. Trousseau, a pathologist, diagnosed his own pancreatic carcinoma on the basis of the association that now bears his name. Hypercoagulable states include hereditary diseases such as deficiencies in antithrombin III, protein C, or protein S; mutation in factor V gene (factor V Leiden) or factor II gene (prothrombin G20210A); and hyperhomocysteinemia. However, a thorough search for identifiable risk factors will come up negative in 25% to 50% of patients with VTE.

DEEP VEIN THROMBOSIS Most DVT starts in the calf veins. Without treatment, 15% to 30% of these clots propagate to the proximal calf veins. The risk for a subsequent PE is much higher with proximal DVT than those confined to the distal calf vessels (40% to 50% versus 5% to 10%, respectively). Involvement of the upper extremities is much less common, but subclavian and axillary vein thrombosis also can lead to PE in as many as 30% of affected individuals. The same risk factors that cause lower extremity DVT also cause upper extremity DVT. In addition, other specific causes of upper extremity DVT include traumatic damage of the vessel intima from heavy exertion such as rowing, wrestling, or weight lifting (PagetSchroetter syndrome); from extrinsic compression at the level of thoracic inlet (thoracic outlet obstruction); or from insertion of central venous catheters or pacemakers. Pain and swelling are the major complaints from patients with DVT; however, a large number of patients with DVT are asymptomatic, particularly if the DVT is restricted to the calf. Patients with upper extremity DVT can develop the superior vena caval syndrome of facial swelling, blurred vision, and dyspnea. Thoracic outlet obstruction can compress the brachial plexus leading to unilateral arm pain associated with hand weakness. Physical examination frequently reveals tenderness, erythema, warmth, and swelling below the site of thrombosis. Pain with dorsiflexion of the foot (Homan sign) may be present, but the low sensitivity and the low specificity limit its usefulness in the diagnosis of lower extremity DVT. A palpable tender cord, dilated superficial veins, and low-grade fever occur in some patients. Upper extremity DVT can cause brachial plexus tenderness in the supraclavicular fossa and atrophic hand muscles. For

171

patients with probable thoracic outlet obstruction, several provocative tests should be performed. Adson test is positive if the radial pulses weaken during inspiration and during extension of the arm of the affected side while rotating the head to the same side. Wright test is positive if the radial pulses become weaker and painful symptoms are reproduced while abducting the shoulder of the affected side with the humerus externally rotated. The laboratory diagnosis of DVT includes measurement of D-dimers, which are fibrin degradation products. D-dimer elevation is a highly sensitive indicator of DVT that can be performed rapidly in the emergency department. In a patient in whom the index of probability is low, a negative D-dimer test effectively excludes the diagnosis of DVT. However, the test is not specific and can be elevated in many other conditions frequently encountered in hospitalized patients (e.g., inflammation, recent surgery, malignancy). Duplex ultrasonography can be used to demonstrate the presence of a blood clot or noncompressibility of the affected veins proximal to the site of occlusion. Duplex ultrasonography has greater sensitivity in detecting proximal DVT (90% to 100%) than distal DVT (40% to 90%) of the lower extremities. With upper extremity DVT, acoustic shadowing of the clavicle may obscure detection of thrombosis in subclavian vein segments. MR angiography is particularly helpful in making the diagnosis of upper extremity DVT and pelvic vein thrombosis. Contrast venography is the conventional gold standard test, but it is invasive and technically difficult in patients with edematous extremities. Therefore, invasive venography should be reserved for patients in whom the clinical suggestion is high, despite negative or inconclusive results from noninvasive imaging. Patients with proximal lower extremity DVT and those with upper extremity DVT should be treated initially with subcutaneous low-molecular-weight heparin (LMWH), intravenous or subcutaneous unfractionated heparin (UFH), or subcutaneous selective factor Xa inhibitor fondaparinux to prevent thrombus propagation and to maintain the patency of venous collaterals. Intravenous UFH should be given as a bolus, followed by continuous infusion to maintain an activated partial thromboplastin time of at least 1.5 times the control value. LMWH and fondaparinux has a longer half-life than UFH and can be given once or twice daily with similar efficacy. Oral warfarin should be initiated together with LMWH, UFH, or fondaparinux without delay and titrated until the international normalized ratio (INR) reaches a value between 2 and 3. When DVT is confined to the calf, the risk for PE is low, and the risk-to-benefit ratio of anticoagulation remains controversial. When upper extremity DVT occurs in young patients who are otherwise healthy, two invasive approaches to thrombus removal should be considered: (1) infusion of a fibrinolytic drug through a catheter inserted directly into the affected vein, or (2) mechanical fragmentation of the thrombus through catheter-based technology. The purpose of these invasive procedures is to prevent or minimize the postthrombotic syndrome, which includes chronic arm pain, swelling, hyperpigmentation, and ulceration from residual venous obstruction. Catheter-based placement of a filter in the inferior vena cava should be considered for patients with proximal DVT who either have an absolute contraindication to anticoagula-

172

Section III—Cardiovascular Disease

tion or develop recurrent PE despite an adequate trial of anticoagulation. Vena cava filters are effective in reducing the incidence of PE, but they increase the risk for recurrent DVT. Some proximal or distal migration of the filter occurs in up to 50% of cases; however, clinically evident filter embolization is limited to case reports.

PULMONARY EMBOLISM PE occurs when a thrombus dislodges from the deep veins of the upper or lower extremities. Pulmonary vascular resistance and pulmonary arterial pressure increase from two mechanisms: (1) anatomic reduction in cross-sectional area of the pulmonary vascular bed, and (2) functional hypoxiainduced pulmonary vasoconstriction. The pressure overload on the right ventricle can lead to dilation, hypokinesis, and tricuspid regurgitation. When severe, elevated right ventricular end-diastolic pressure can compress the right coronary artery, causing subendocardial ischemia. In acute PE, areas of lung tissue are ventilated but underperfused. This mismatch and the resultant redistribution of pulmoV Q nary blood flow from obstructed pulmonary artery to other cause arterial hypoxemia. In lung regions with lower V Q patients with a patent foramen ovale, hypoxemia worsens when the sudden elevation in right atrial pressure causes right-to-left shunting across the foramen. The classic symptoms of acute PE are the sudden onset of dyspnea and pleuritic chest pain. Additional symptoms include anginal chest pain from right ventricular ischemia, hemoptysis from pulmonary infarction, and syncope or presyncope from massive PE with acute right ventricular failure (cor pulmonale). The most common physical findings are tachypnea and tachycardia. Additional physical findings include a right ventricular lift, inspiratory crackles, a loud pulmonary component of the second sound, expiratory wheezing, and a pleural rub. Symptoms and signs of proximal DVT are present in 10% to 20% of patients. Arterial blood gas analysis often reveals hypoxemia, respiratory alkalosis, and a high alveolar-to-arterial oxygen tension gradient. However, normal arterial blood gas values do not exclude the diagnosis. The most common finding with ECG analysis is sinus tachycardia. Atrial fibrillation, premature atrial contraction, and supraventricular tachycardia are less common. Other ECG changes suggest acute right ventricular strain. These include the S1-Q3-T3 pattern, a new right bundle branch block or right axis deviation, and P-wave pulmonale. However, these findings are present in only 30% of patients with even massive PE. Common but nonspecific abnormalities with chest radiographic studies include atelectasis, pleural effusion, and pulmonary infiltrates. Less common but more specific radiographic findings include Hampton hump (i.e., wedge-shaped infiltrate in the peripheral lung field), which is indicative of pulmonary infarction, and Westermark sign (decreased vascularity). The plasma D-dimer test is elevated in most patients with PE as a result of activation of the endogenous fibrinolytic system, which is not sufficient to dissolve the clot. Commercially available D-dimer assays have a high sensitivity and negative predictive value but low specificity. Therefore, a normal D-dimer test effectively excludes the diagnosis of PE in patients in whom the clinical suggestion is low or intermediate. However, it should not be used to screen patients with a high

Figure 13-5 Spiral chest computed tomographic angiogram shows a large thrombus in the right main pulmonary artery (arrow). (Image provided by Michael Landay, MD, Department of Radiology, University of Texas Southwestern Medical Center, Dallas, Texas.)

index of suspicion because of low negative predictive value. Elevated levels of cardiac troponin I and troponin T and other markers of myocardial injury can be found in patients with PE and are indicative of right ventricular dysfunction and a poor prognosis. In patients with suggested PE, a completely normal V Q scan effectively excludes the diagnosis without further test scans are interpreted ing. However, less than 10% of V Q as definitively normal. In patients in whom a moderate or high level of clinical probability of PE exists, a high scan has a diagnostic accuracy of 90% to probability V Q 100%; however, a low or intermediate probability scan is no more helpful than a coin flip. More recently, multidetector CT angiography has become the imaging modality of choice in patients with acute PE because of its excellent visualization of the pulmonary artery (Fig. 13-5). The resolution of 1mm or less rivals that of conventional invasive angiography. The speed of the newer-generation scanners allows acquisition of all images within a single breath-hold, avoiding respiratory motion artifacts. The overall negative predictive value of multidetector CT angiography exceeds 99%. A negative CT scan excludes the diagnosis of PE and eliminates the need for further diagnostic testing. CT also permits detection of other pathologic conditions involving the lung parenchyma, pleura, and mediastinal structures. Such pathologic findings may mimic PE and constitute alternative causes of chest pain and dyspnea. Multidetector CT angiography is not yet available at all centers. The requirement for intravenous injection of iodinated contrast material restricts applicability to those without a history of kidney disease or an allergic reaction to contrast dye. Figure 13-6 presents an algorithm for the work-up of PE based on current evidence. Echocardiography may directly detect thrombi in the right atrium, right ventricle, or pulmonary artery or indirectly demonstrate right ventricular dysfunction, signifying presence of hemodynamically significant emboli. Therefore, it is helpful in diagnosis of PE in patients with hypotension or shock, particularly when multidetector CT are not immediately available. Invasive pulmonary angiography should

Chapter 13—Vascular Diseases and Hypertension Pre-test Probability Score Clinical signs of DVT Heart rate >100 beats/min Recent surgery or immobilization Previous DVT or PE Hemoptysis Cancer PE more likely than alternative diagnosis

3 1.5 1.5 1.5 1 1 3

Score 30mg/g New elevation in serum creatinine, significant elevation in serum creatinine with initiation of ACEI or ARBs, refractory hypertension, flash pulmonary edema, abdominal bruit Arm pulses > leg pulses, arm BP > leg BP, chest bruits, rib notching on chest radiograph Hypokalemia, refractory hypertension

Truncal obesity, wide and blanching purple striae, muscle weakness Spells of paroxysmal hypertension, palpitations, perspiration, pallor, pain in the head Diabetes Loud snoring, daytime somnolence, obesity, large neck

MR or CT angiography, invasive angiogram Chest MRI or CT, aortogram Plasma renin and aldosterone, 24-hr urine potassium, 24-hr urine aldosterone and potassium after salt loading, adrenal CT, adrenal vein sampling 24-hr urine cortisol, dexamethasone suppression test, adrenal CT Plasma and 24-hr urine metanephrines and catecholamines, adrenal CT

Sleep study

ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; BP, blood pressure; CT, computed tomography; GFR, glomerular filtration rate; MR, magnetic resonance; MRI, magnetic resonance imaging. Data from Kaplan NM: Clinical Hypertension, 8th ed. Philadelphia, Williams & Wilkins, 2002.

pelling cause is found on the initial evaluation, or (2) when the hypertensive process is so severe that it either is refractory to intensive multiple-drug therapy or requires hospitalization. Table 13-6 summarizes the major causes of secondary hypertension that should be suggested on the basis of a good history, physical examination, and routine laboratory tests.

Renal Parenchymal Hypertension Chronic kidney disease is the most common cause of secondary hypertension. Hypertension is present in more than 85% of patients with chronic kidney disease and is a major factor causing their increased cardiovascular morbidity and mortality. The mechanisms causing the hypertension include an expanded plasma volume and peripheral vasoconstriction, with the latter caused by both activation of vasoconstrictor pathways (renin-angiotensin and sympathetic nervous systems) and inhibition of vasodilator pathways (nitric oxide). Renal insufficiency should be considered when proteinuria is found by dipstick or when the serum creatinine level is greater than 1.2 mg/dL in women with hypertension or greater than 1.4 mg/dL in men with hypertension.

Renovascular Hypertension Unilateral or bilateral renal artery stenosis is present in less than 2% of patients with hypertension in a general medical practice but in up to 30% in patients with medically refractory hypertension. The main causes of renal artery stenosis are atherosclerosis (85% of patients), typically in older adults with other clinical manifestations of systemic atherosclerosis, and fibromuscular dysplasia (15% of patients), typically in women between the ages of 15 and 50 years. Unilateral renal artery stenosis leads to underperfusion of the juxta-

glomerular cells, thereby producing renin-dependent hypertension even though the contralateral kidney is able to maintain normal blood volume. In contrast, bilateral renal artery stenosis (or unilateral stenosis with a solitary kidney) constitutes a potentially reversible cause of progressive renal failure and volume-dependent hypertension. The following clinical clues increase the suggestion of renovascular hypertension: any hospitalization for urgent or emergent hypertension; recurrent flash pulmonary edema; recent worsening of longstanding, previously well-controlled hypertension; severe hypertension in a young adult or in an adult after 50 years of age; precipitously and progressively worsening of renal function in response to angiotensin-converting enzyme (ACE) inhibition or angiotensin receptor blockade (ARB); unilateral small kidney by any radiographic study; extensive peripheral arteriosclerosis; or a flank bruit. The diagnosis is confirmed by noninvasive testing with MR or spiral CT angiography (Fig. 13-11). Renal artery angioplasty often cures fibromuscular dysplasia. Atherosclerotic renal artery stenosis should be treated with intensive medical management of atherosclerotic risk factors (hypertension, lipids, smoking cessation). Revascularization should be considered for the following indications: (1) medically refractory hypertension, (2) progressive renal failure on medical therapy, and (3) bilateral renal artery stenosis or stenosis of a solitary functioning kidney.

Primary Aldosteronism The most common causes of primary aldosteronism are (1) a unilateral aldosterone-producing adenoma and (2) bilateral adrenal hyperplasia. Because aldosterone is the principal ligand for the mineralocorticoid receptor in the distal nephron, excessive aldosterone production causes excessive renal Na−-K− exchange, often resulting in hypokalemia. The

178

Section III—Cardiovascular Disease

“String of beads” Proximal stenosis

Figure 13-11 Computed tomographic angiogram with three-dimensional reconstruction. A, Classic string-of-beads lesion of fibromuscular dysplasia. B, Severe proximal atherosclerotic stenosis of the right renal artery. (Images courtesy of Bart Domatch, MD, Radiology Department, University of Texas Southwestern Medical Center, Dallas, Texas.)

diagnosis should always be suggested when hypertension is accompanied by either unprovoked hypokalemia (serum K+ < 3.5mmol/L in the absence of diuretic therapy) or a tendency to develop excessive hypokalemia during diuretic therapy (serum K+ < 3 mmol/L). However, more than one third of patients do not have hypokalemia on initial presentation, and the diagnosis should be considered in any patient with refractory hypertension. The diagnosis is confirmed by the demonstration of nonsuppressible hyperaldosteronism during salt loading, followed by adrenal vein sampling to distinguish between a unilateral adenoma and bilateral hyperplasia. Laparoscopic adrenalectomy is the treatment of choice for unilateral aldosterone-producing adenoma, whereas pharmacologic mineralocorticoid receptor blockade with eplerenone is the treatment for bilateral adrenal hyperplasia.

Mendelian Forms of Hypertension Nine rare forms of severe early-onset hypertension are inherited as mendelian traits. In each case, the hypertension is mineralocorticoid induced and involves excessive activation of the epithelial sodium channel (ENaC), the final common pathway for reabsorption of sodium from the distal nephron. The resultant salt-dependent hypertension can be caused by gain-of-function mutations of ENaC (Liddle syndrome) or the mineralocorticoid receptor (i.e., a rare form of pregnancy-induced hypertension) and by increased production or decreased clearance of mineralocor ticoids. These include aldosterone (glucocorticoid-remediable aldosteronism), deoxycorticosterone (17-hydroxylase deficiency), and cortisol (syndrome of apparent mineralo corticoid excess).

Pheochromocytoma Pheochromocytomas are rare catecholamine-producing tumors of the adrenal (or sometimes extra-adrenal) chromaf fin cells. The diagnosis should be suggested when hyperten-

sion is accompanied by paroxysms of headaches, palpitations, pallor, or diaphoresis. In some patients, pheochromocytoma is misdiagnosed as panic disorder. A family history of earlyonset hypertension may suggest pheochromocytoma as part of the multiple endocrine neoplasia syndromes. If the diagnosis is missed, outpouring of catecholamines from the tumor can cause an unsuspected hypertensive crisis during unrelated radiologic or surgical procedures; the perioperative mortality rate exceeds 80% in such patients. Laboratory confirmation of pheochromocytoma is made by demonstrating elevated levels of plasma or urinary metanephrines, catecholamines, or other metabolites such as vanillylmandelic acid. These are typically large tumors that can usually be localized by CT or magnetic resonance imaging (MRI), although nuclear scanning with specific isotopes that localize to chromaffin tissue is occasionally needed to identify smaller tumors. Treatment of these tumors is surgical resection. Patients must receive adequate α blockade (phentolamine), β blockade, and volume expansion before surgery to prevent the hemodynamic swings that can occur during manual manipulation of the tumor peri-operatively. For unresectable tumors, chronic therapy with the α-adrenergic blocker phenoxybenzamine is usually effective. The differential diagnosis includes other causes of neurogenic hypertension such as sympathomimetic agents (cocaine, methamphetamine), baroreflex failure, and obstructive sleep apnea. A history of surgery and radiation therapy for head-and-neck tumors suggests the possibility of baroreceptor damage. Loud snoring, obesity, and somnolence suggest obstructive sleep apnea. Weight loss, continuous positive airway pressure, and corrective surgery improve BP control in some patients with sleep apnea. Other causes of secondary hypertension include hypo thyroidism, hyperthyroidism coarctation of the aorta, and immunosuppressive drugs, especially cyclosporine and tacrolimus.

Chapter 13—Vascular Diseases and Hypertension

TREATMENT OF HYPERTENSION Prescription medication is the cornerstone of treating hypertension. Lifestyle modification should be used as an adjunct but not as an alternative to life-saving BP medication. Most dietary sodium comes from processed foods, and daily salt consumption can be reduced from 10 to 6 g by teaching patients to read food labels (6 g of NaCl = 2.4 g of Na+ = 100mmol of Na+. The Dietary Approaches to Stop Hypertension (DASH) guidelines, which are rich in fresh fruits and vegetables (for high potassium content) and low-fat dairy products, has been shown to lower BP in feeding trials. Other lifestyle modifications that can lower BP include weight loss in overweight patients with hypertension, regular aerobic exercise, smoking cessation, and moderation in alcohol intake. Currently, 92 prescription medications and many fixeddose combinations are marketed for the treatment of hypertension in the United States (see Table 13-4).

WHICH DRUGS FOR WHICH PATIENTS? Patients with Uncomplicated Hypertension Choosing the best drugs to treat hypertension in a given patient comes down to two considerations: (1) effectively lowering BP and preventing hypertensive complications with minimal side effects and cost, and (2) concomitant treatment of co-morbid cardiovascular diseases (e.g., angina, heart failure). The seventh report of the U.S. Joint National Committee (JNC 7) recommends a thiazide-type diuretic as cost-effective first-line therapy for most patients with hypertension. It also recommends initiating therapy with two drugs—one being a thiazide—for stage 2 hypertension. In contrast, the European Society of Hypertension makes no specific drug class recommendation, arguing that the most effective drugs are those that the patient will tolerate and take. The British Hypertension Society advocates a treatment strategy that is based on the patient’s age and ethnicity. It recommends initiating therapy with an ACE inhibitor or ARB or a β blocker (A or B drug) for young white patients (PV

PO2

100

B Figure 16-9 A rising Pco2 leads to a linear increase in minute ventilation (A). The ventilatory response to hypoxemia (B) is less sensitive and is clinically relevant only when the Po2 has dropped significantly.

receptor located between the internal and external branches of carotid artery. Changes in Pao2 are sensed by the carotid sinus nerve. Neural traffic projects to the respiratory center through the glossopharyngeal nerve, which serves to modulate ventilation. The carotid body also senses changes in Paco2 and pH. Nonvolatile acids (i.e., ketoacids) stimulate ventilation through their effects on the carotid body. The outcome of this complex respiratory control system is that variables such as Pao2, Paco2, and pH are held within narrow limits under most circumstances. The respiratory control center also can adjust VT and f to minimize the energetic cost of breathing and can adapt to special circumstances such as speaking, swimming, eating, and exercise. Breathing can be stimulated when Pco2, Po2, and pH are artificially manipulated. For example, rebreathing carbon dioxide, inhaling a concentration of low oxygen, or infusion of acid into the bloodstream will increase ventilation.

PERFUSION The pulmonary vascular bed differs from the system circulation in several respects. The pulmonary vascular bed receives

the entire cardiac output of the right ventricle, whereas the cardiac output from the left ventricle is dispersed among several organ systems. Despite receiving the entire cardiac output, the pulmonary system is a low-resistance, lowpressure circuit. The normal mean systemic arterial pressure is about 100mmHg, whereas the normal mean pulmonary artery pressure is in the range of 15mmHg. The vasculature bed can passively accommodate an increase in blood flow without raising arterial pressure by recruiting more vessels in the lung. Thus, during exercise, there is little increase in pulmonary artery resistance despite a large increase in pulmonary blood flow. Hypoxic vasoconstriction also is a feature unique to the pulmonary vascular system and regulates regional blood flow. This regulation aids in matching blood flow to ventilation by reducing flow to poorly ventilated regions of the lung. In the upright individual, there is greater perfusion of the lung bases than apices (Fig. 16-10). In a low-pressure system such as the pulmonary circulation, the effects of gravity on blood flow need to be considered. Usually the arterialvenous pressure difference provides the “driving” pressure for blood flow. Although this is true for the systemic circulation, it is true only for certain regions of the lung. With the respiratory system, pulmonary blood flow also needs to be considered in the context of alveolar pressure. Venous and arterial pressures are importantly affected by gravity, whereas alveolar pressure remains constant throughout the lung, assuming the airways are open. Thus, as one descends from the apex to the base of the lung, arterial and venous pressures increase because of gravity, but alveolar pressure remains constant. At the apex of the lung, alveolar pressure may be greater than arterial pressure. This region of the lung is referred to as zone 1 and, in theory, receives no blood flow. Alveolar pressure may exceed arterial pressure under special circumstances such as hypovolemic shock, in which pulmonary arterial pressure may fall below alveolar pressure, and with very high levels of positive end-expiratory pressure

Section IV—Pulmonary and Critical Care Medicine

GAS TRANSFER Oxygen and carbon dioxide are easily dissolved in plasma. Nitrogen is much less soluble and is not significantly exchanged across the alveolar-capillary interface. The pressure gradient for oxygen between the alveolus and capillary promotes diffusion of oxygen from the alveolus to the capillary (Pao2 of 150mmHg versus Pao2 of 40mmHg). This pressure difference is greater than that driving carbon dioxide from the mixed venous blood to the alveolus (Pmvco2 of 45mmHg versus Paco2 of 40mmHg). Despite the lower driving pressure, the greater solubility of carbon dioxide allows complete equilibration between the alveolus and plasma during each respiratory cycle (Fig. 16-11). Most of the oxygen contained in the blood is bound to hemoglobin, with a small fraction dissolved and measured as the Pao2. The amount of oxygen dissolved is about 3mL/L in arterial blood, whereas the amount of oxygen bound to hemoglobin is about 197mL/L in arterial blood, assuming a normal hematocrit. Each molecule of hemoglobin is capable of carrying four molecules of oxygen. The shape of the oxyhemoglobin association curve reflects the cooperative binding of oxygen to hemoglobin (Fig. 16-12). In general, the percent hemoglobin saturation is between 80% and 100% with Pao2 above 60mmHg and drops dramatically when the Pao2 is less than 60 mm Hg. Factors that decrease the affinity of hemoglobin for oxygen include a reduction in blood pH, an increase in temperature, an increase in Pco2, and an increase in 2,3-diphosphoglyceric acid (2,3-DPG). These factors facilitate unloading of oxygen into tissues. This is seen as a shift of the oxyhemoglobin dissociation curve to the right. The oxygen carrying capacity of hemoglobin is also affected by competitive inhibitors for binding sites such as carbon monoxide. Carbon monoxide has an affinity for hemoglobin that is 240 times greater than oxygen. Thus, it will preferentially bind to hemoglobin. However, it does not affect the amount of oxygen dissolved in the blood. Someone with carbon monoxide poisoning

Pul. capillary

Pul. artery

Pul. vein

Alv PO2 = 100 mm Hg

100

PO2 (mm Hg)

(PEEP), which may increase alveolar pressure to the extent at which it becomes greater than arterial pressure. As one descends from the apex toward the midzone of the lung, arterial and venous pressures increase, and alveolar pressure remains constant. At some point, arterial pressure becomes greater than alveolar pressure. In this region, the driving pressure for blood flow is the arterial-alveolar pressure difference. This region is referred to as zone 2 of the lung. Normally, there is very little zone 2 because alveolar pressure is less than venous pressure in most of the lung. However, with high levels of PEEP, alveolar pressure will become greater than venous pressure in more lung regions. When further approaching the base of the lung, the effects of gravity on arterial and venous pressures are more pronounced, venous pressure becomes greater than alveolar pressure, and the arterial-venous pressure difference provides the driving pressure for blood flow. This region is referred to as zone 3 of the lung. Normally, most of the lung is in zone 3, and most of the perfusion is to the lung base. This inequality in perfusion from apex to base is qualitatively similar to the inequality of ventilation from apex to base, thereby optimizing the matching of ventilation and perfusion.

80 60 40 0 50

PcO2 (mm Hg)

204

45 40 0

Alv PcO2 = 40 mm Hg 0

0.25 0.50 0.75 Time (sec.) Figure 16-11 Changes in Po2 and Pco2 as blood courses from the pulmonary artery through the capillaries and into the pulmonary veins. The diffusion gradient is greater for O2 than CO2. However, equilibration of capillary and alveolar gas occurs for both molecules within the 0.75 second it takes for blood to traverse the capillaries.

may have a normal Pao2 but very low blood oxygen content due to the desaturated hemoglobin. About 5% of carbon dioxide is dissolved in plasma, and the remainder is transported in other forms. A small amount of carbon dioxide also binds to hemoglobin. However, carbon dioxide does not exhibit cooperative binding; therefore, the shape of the carbon dioxide–hemoglobin dissociation curve is linear (see Fig. 16-12). Carbon dioxide binds to the protein component of the hemoglobin molecule and to the amino groups of the polypeptide chains of plasma proteins to form carbamino compounds. About 10% of carbon dioxide is transported in this fashion. Most of the carbon dioxide is transported as bicarbonate ion. As carbon dioxide diffuses from metabolically active tissue into the blood, it reacts with water to form carbonic acid. This reaction primarily occurs in the red blood cell because it is catalyzed by the enzyme carbonic anhydrase, which resides in the red blood cell. Carbonic acid then dissociates to bicarbonate and hydrogen ion. Although there is more carbon dioxide dissolved in blood than oxygen, it is still a small fraction of the total carbon dioxide transported by blood.

ABNORMALITIES OF PULMONARY GAS EXCHANGE The Pao2 and Paco2 are determined by the degree of equilibration between the alveolar gas and capillary blood. The degree of equilibration depends on four major factors: (1) ventilation, (2) matching of ventilation with perfusion, (3) shunt, and (4) diffusion. A fifth cause of hypoxemia is a low inspired Po2. Hypoxemia refers to a reduction in the oxygen content in the blood. Specifically, hypoxemia is determined by measuring the Po2 of arterial blood. In contrast, hypoxia

Chapter 16—Evaluating Lung Structure and Function

Increased PCO2, 2,3–DPG, temp. decreased pH

12 8 4 0

60 PO2 (mm Hg)

100

CO2 Content (mL/dL)

A 60

Decreased PO2 Increased PO2

40 PCO2 (mm Hg)

B Figure 16-12 A, The oxyhemoglobin dissociation curve. The bulk of the oxygen is combined with hemoglobin. The various factors that decrease the hemoglobin oxygen affinity are shown. Opposite changes increase hemoglobin oxygen affinity, shifting the curve to the left. B, The carbon dioxide dissociation curve is more linear than the oxyhemoglobin curve throughout the physiologic range. Increased Pao2 shifts the curve to the right, which decreases carbon dioxide content for any given Paco2 and thus facilitates carbon dioxide off-loading in the lungs. The shift to the left at a lower Pao2 facilitates carbon dioxide on-loading at the tissues. DPG, diphosphoglycerate.

refers to a decrease in oxygen content of an organ, for example, myocardial hypoxemia. Hypoventilation is defined as ventilation inadequate to keep Pco2 from increasing above normal. In this situation, hypoxemia may occur when increased carbon dioxide in alveoli displaces alveolar oxygen. The reciprocal relationship between alveolar carbon dioxide and alveolar oxygen is described by the alveolar gas equation. PAO2 = [(PB − PH2 O ) × FIO2 ] − (PaCO2 R ) , where Pao2 is the partial pressure of oxygen in the alveolus; PB is barometric pressure; Ph2o is water vapor pressure; Fio2 is the fraction of inspired oxygen; and R is the respiratory exchange ratio.

(mm Hg)

and PCO2

Normal

From this equation, it is apparent that as alveolar ventilation falls and Paco2 rises, Pao2 will have to fall. Administering supplemental oxygen (increasing the Fio2) can reverse hypoventilation-induced hypoxemia. When breathing room air, the difference between alveolar oxygen and arterial oxygen (Pa-a gradient) is normally about 20mmHg. Generally, this difference increases when hypoxemia is present. However, when hypoxemia is due to hypo ventilation, the O2 gradient is within normal limits. Causes of hypoventilation are varied and range from diseases or drugs that depress the respiratory control center to disorders of the chest wall or respiratory muscles that impair respiratory pump function. Disorders associated with hypoventilation include inflammation trauma or hemorrhage in the brainstem, spinal cord pathology, anterior horn cell disease, peripheral neuropathies, myopathies, abnormalities of the chest wall such as kyphoscoliosis, and upper airway obstruction. Administering a higher Fio2 will alleviate the hypoxemia but will do little to improve the elevated Paco2. A second cause of hypoxemia is ventilation-perfusion ) mismatch. This is the most common cause of ratio ( V Q hypoxemia in disease states. In the ideal lung, ventilation and perfusion would be perfectly matched. Although both ventilation and perfusion are greater at the base relative to is lower at the base than at the apex of the lung, the V Q ranges the apex of the lung. In the normal lung, the V Q from 0.5 at the base to 3 at the apex. The overall V Q for the normal lung is 0.8. If lung disease develops, V Q is less than 0.8, the inequality may develop. If the V Q Pa-a gradient is increased, and hypoxia ensues. The Paco2 is usually within the normal range but will increase slightly at extremely low ratios (Fig. 16-13). Typically, hypoxemia seen in diseases that affect the airways, such as chronic misobstructive pulmonary disease (COPD), is due to V Q match. As with hypoxemia due to hypoventilation, admin istering a higher Fio2 improves hypoxemia due to V Q mismatch.

100 80 60 40 20

PO2

O2 Content (mL/dL)

205

Normal Moderate Severe V ˙ A/Q˙ Inequality

Figure 16-13 A, The effect of increasing ventilation-perfusion ( V Q ) inequality on Pao2 and Paco2 when cardiac output and minute ventilation are held constant. B, The gas tensions change when ventilation is allowed to increase. Increased ventilation can maintain a normal Paco2 but can only partially correct the hypoxemia. (Adapted from Dantzker DR: Gas exchange abnormalities. In Montenegro H [ed]: Chronic Obstructive Pulmonary Disease. New York, Churchill Livingstone, 1984, pp 141-160.)

Section IV—Pulmonary and Critical Care Medicine

QS Qt = (CcO2 − CaO2 ) ( CcO2 − CvO2 ) , where QS is the shunted blood flow, Qt is the total blood flow, Cco2 is the end pulmonary capillary oxygen content; Cao2 is the arterial oxygen content; and Cvo2 is the mixed venous oxygen content. If the shunt is severe enough, patients will require mechanical ventilation and the application of positive end expiratory pressure to improve arterial oxygenation. At values less than 50% of the cardiac output, a shunt has very little effect on Paco2 (Fig. 16-14). With shunt, the Pa-a gradient is elevated, and the Paco2 is within normal range or may be low. Unlike mismatch, hypoxemia due to hypoventilation or V Q administering a high Fio2 does not improve hypoxemia due to shunt. The fourth cause of hypoxemia is diffusion impairment. With normal cardiopulmonary function, the blood spends on average 0.75 second in the pulmonary capillaries. Typically, it only takes 0.25 second for the alveolar oxygen to diffuse across the thin alveolar capillary membrane and equilibrate with pulmonary arterial blood. However, if there is impairment to diffusion across this membrane (thickening of the alveolar capillary membrane by fluid, fibrous tissue, cellular debris, or inflammatory cells), it will take longer for the oxygen in the alveoli to equilibrate with pulmonary arterial blood. If the impediment to diffusion is such that it takes longer than 0.75 second for oxygen to diffuse, hypoxemia ensues. Alternatively, if the time the red cell spends traversing the pulmonary capillary decreases to 0.25 second or less, hypoxemia may develop. Hypoxemia may only be evident during exercise in individuals with diffusion

(mm Hg)

100

60 and PCO2

The third cause of hypoxemia is shunt. A right-to-left shunt occurs when a portion of blood travels from the right side to the left side of the heart without the opportunity to exchange oxygen and carbon dioxide in the lung. Right to left shunts can be classified as anatomic or physiologic. With an anatomic shunt, a portion of the blood bypasses the lung by traversing through an anatomic canal. In all healthy individuals, there is a small fraction of blood in the bronchial circulation that passes to the pulmonary veins and empties into the left atrium, thereby reducing Pao2 of the systemic circulation. A smaller portion of the normal shunt is related to the coronary circulation draining through the thebesian veins into the left ventricle. Anatomic shunts found in disease states can be classified as intracardiac or intrapulmonary shunts. Intracardiac shunts occur when right atrial pressures are elevated and deoxygenated blood travels from the right atrium to the left atrium through an atrial septal defect or patent foramen ovale. Intrapulmonary anatomic shunts consist primarily of arteriovenous malformations or telangiectasias. A physiologic right-to-left shunt consists of a portion of the pulmonary arterial blood passing through normal vasculature but not coming into contact with alveolar air. This is an extreme example of ventilation-perfusion = 0). Physiologic shunt can be due to mismatch ( V Q diffuse flooding of the alveoli with fluid, as seen with congestive heart failure or acute respiratory distress syndrome. Alveolar flooding with inflammatory exudates as seen in lobar pneumonia also causes a shunt. The fraction of blood shunted can be calculated when the Fio2 is 100% by using the following equation.

PO2

206

Percent shunt Figure 16-14 The effect of increasing shunt on the arterial Pao2 and Paco2. The minute ventilation has been held constant in this example. Under usual circumstances, the hypoxemia would lead to increased minute ventilation and a fall in the Paco2 as the shunt increases. (From Dantzker DR: Gas exchange abnormalities. In Montenegro H [ed]: Chronic Obstructive Pulmonary Disease. New York, Churchill Livingstone, 1984, pp 141-160.)

impairment because of the shortened red cell transit time. In this instance, the O2 gradient may be normal at rest but increases with exercise. With diffusion impairment, the Paco2 generally is within the normal range. As with hypox mismatch, adminisemia due to hypoventilation and V Q tering a higher Fio2 improves hypoxemia due to impaired diffusion. A fifth cause of hypoxemia is due to low inspired oxygen. This is seen at an altitude at which the fraction of inspired oxygen is normal, but the partial pressure of oxygen is low because barometric pressure is low (Patm). Rarely, circumstances occur in which the Fio2 is low (e.g., rebreathing air). Hypoxemia due to low inspired oxygen is associated with a normal O2 gradient and is usually accompanied by a low Paco2. Providing supplemental oxygen will correct this form of hypoxemia. Finally, a low mixed venous Po2 will predispose individuals to hypoxia (Fig. 16-15).

Evaluation of Lung Function Pulmonary function tests evaluate one or more major aspects of the respiratory systems. Accurate measurements of lung volumes, airway function, and gas exchange require a pulmonary function testing laboratory. Pulmonary function tests are commonly used to aid in the diagnosis of disease and assess disease severity. In addition, they are helpful in monitoring the course of the disease, assessing risk for surgical procedures, and measuring the effects of varied environmental exposures. Assessment of bronchodilator response or other forms of treatment also can be evaluated with serial pulmonary function tests (Table 16-1). Accurate interpretation of pulmonary function tests requires the appropriate reference standards. Variables that affect the predicted

Chapter 16—Evaluating Lung Structure and Function Inspiratory reserve volume

Arterial oxygen content (mL/dL)

22 Normal lung

21 20

Total lung capacity

40% Shunt

Expiratory reserve volume Functional residual capacity Residual volume

18 17 16 15

30 40 Mixed venous PO2 (mm Hg)

Vital capacity

Tidal volume

Va/Q Inequality

207

Figure 16-15 The effect of altering mixed venous PVo2 on the arterial oxygen content under three assumed conditions: a ) inequality, and normal lung, severe ventilation-perfusion ( V Q the presence of a 40% shunt. For each situation, the patient is breathing 50% oxygen and the PVo2 or mixed venous Po2 is altered, keeping all other variables constant. (From Dantzker DR: Gas exchange in the adult respiratory distress syndrome. Clin Chest Med 3:57-67, 1982.)

Figure 16-16 Lung volumes and capacities. Although spirometry can measure vital capacity and its subdivisions, calculation of residual volume requires measurement of functional residual capacity by body plethysmography, helium dilution technique, or nitrogen washout.

Table 16-1 Indications for Pulmonary Function Testing

standards for pulmonary function tests include age, height, gender, race, and hemoglobin concentration. Spirometry is the simplest means of measuring lung function and can be performed in an office practice. A spirometer is an apparatus that measures inspiratory and expiratory volumes. Flow rates can be calculated from tracings of volume versus time. Typically, vital capacity (VC) is measured as the difference between a full inspiration to total lung capacity (TLC) and a full exhalation to residual volume (RV) (Fig. 16-16). Flow rates are measured after instructing someone to forcefully exhale from TLC to RV. Such a forced expiratory maneuver allows one to calculate the forced expired volume in 1 second (FEV1) and the forced vital capacity (FVC) (Fig. 16-17). A value of 80% to 120% predicted is considered normal for FVC. Normally, people can exhale more than 75% to 80% of their FVC in the first second, and the majority of FVC can be exhaled in 3 seconds. The ratio of these two variables is normally more than 0.8.

Volume

FEV1

FVC

MMEF or FEF25–75 0

2 Time (sec) Normal

FVC

FEV1 Volume

• Evaluation of signs and symptoms • Shortness of breath, exertional dyspnea, chronic cough • Screening at-risk populations • Monitoring pulmonary drug toxicity • Follow-up abnormal study • Chest radiograph, electrocardiogram, arterial blood gases, hemoglobin • Preoperative assessment • Assess severity • Follow response to therapy • Determine further treatment goals • Assess disability

MMEF or FEF25–75 0

2 3 4 5 Time (sec) Obstructed Figure 16-17 Spirometry in a normal individual and in a patient with obstructive lung disease. FEV1 represents the forced expired volume in 1 second, and FVC represents the forced vital capacity. The slope of the line connecting the points at 25% and 75% of the FVC represents the forced expired flow (FEF at 25% to 75%) or maximum mid-expiratory flow (MMEF). The FEF at 25% to 75% is less reproducible and less specific than the FEV1.

208

Section IV—Pulmonary and Critical Care Medicine

Spirometry can reveal abnormalities that are classified into two patterns—obstructive and restrictive. Obstructive impairments are defined by a low FEV1/FVC ratio. Diseases characterized by an obstructive pattern include asthma, chronic bronchitis, emphysema, bronchiectasis, cystic fibrosis, and some central airway lesions. The reduction in FEV1 (expressed as % predicted FEV1) is used to determine the severity of airflow obstruction (Web Fig. 16-2). Peak expiratory flow rate (PEFR) can be measured as the maximal expired flow rate obtained during spirometry or when using a hand-held peak flow meter. The lower the peak expiratory flow rate, the more significant the obstruction. The peak flow meter can be used at home or in the emergency department and to evaluate the presence of obstruction. Severe attacks of asthma, for example, are usually associated with peak expiratory flow rates of less than 200 L per minute (normal is 500 to 600L per minute). A restrictive pattern is characterized by loss of lung volume. With spirometry, both the FVC and FEV1 are reduced with a normal FEV1/FVC ratio. The restrictive pattern must be confirmed by measurements of lung volumes. Lung volumes are measured using body plethysmography or by dilution of an inert gas such as helium. Lung volumes that can be measured with these techniques include FRC, TLC, and RV (see Fig. 16-16). As mentioned previously, FRC is the lung volume at which the inward elastic recoil of the lung equals the outward elastic recoil of the chest wall. Changes in FRC reflect abnormalities in lung elastic recoil. Diseases associated with increased elastic recoil such as pulmonary fibrosis are associated with a reduction in FRC, whereas those with decreased recoil such as emphysema are associated with an increase in FRC. TLC is the amount of air in the thorax after a maximal inspiration and is determined by the balance of the forces generated by the respiratory muscles to expand the respiratory system and the elastic

PEF

recoil of the respiratory system. Restrictive lung disease is defined as a TLC less than 80% predicted, whereas values of TLC greater than 120% predicted are consistent with hyperinflation. The lower the percent predicted TLC, the more severe the restrictive impairment. Restriction may be due to disorders of the lung, chest wall, respiratory muscles, or pleural space. Lung diseases that cause pulmonary fibrosis will cause a restrictive pattern because of the increased elastic recoil of the respiratory system. Diseases of the chest wall, such as kyphoscoliosis, obesity, and ankylosing spondylitis can also cause restriction by reducing the elasticity of the chest wall. Weakness of the respiratory muscles causes restriction by reducing the force available to inflate the respiratory system. Myasthenia gravis, amyotrophic lateral sclerosis, diaphragm paralysis, and Guillain-Barré syndrome can be associated with weakness sufficient to cause restrictive lung disease. Finally, space-occupying lesions involving the pleural space such as pleural effusions, pneumothorax, or pleural tumors can cause restriction. Occasionally, RV and FRC may be elevated with no increase in total lung capacity. This pattern is referred to air trapping and can be seen with COPD or asthma. The forced expiratory maneuver can be analyzed in terms of flow and volume, that is, a flow-volume loop (Fig. 16-18). Flow-volume loops are useful when identifying obstructive and restrictive patterns. The characteristic appearance of obstructive impairment is concavity (“scooping”) of the expiratory loop (Web Fig. 16-3). With restrictive impairments, the loops are similar in appearance to normal but reduced in size. In addition, flow-volume loops are the primary means of identifying upper airway obstruction. Upper airway obstruction is characterized by a truncated (clipped) inspiratory or expiratory loop. A fixed obstruction has clipping of both inspiratory and expiratory loops. Variable intrathoracic upper airway obstruction exhibits

Expiration

Normal UAO

OLD

FEF75

PIF

RLD

Flow

FEF50

Inspiration Volume

Volume

Figure 16-18 A, The maximum expired flow and volume curve in a normal individual. The peak expiratory flow (PEF) and forced expiratory flows at 50% and 75% of the exhaled vital capacity (FEF at 50% and 75%) are indicated. PIF, peak inspiratory flow. B, In obstructive lung disease (OLD), hyperinflation pushes the position of the curve to the left, and characteristic scalloping on expiration develops. In restrictive lung disease (RLD), lung volumes are reduced, but flow for any point in volume is normal. The flow-volume curve displays different patterns with various forms of upper airway obstruction (UAO), with reduction in respiratory flow if the obstruction is outside the thoracic cavity and, in addition, in expiratory flow if the obstruction is caused by a fixed deformity.

Chapter 16—Evaluating Lung Structure and Function clipping of the expiratory loop, whereas variable extrathoracic obstruction exhibits clipping of the inspiratory loop (see Fig. 16-18).

BRONCHOPROVOCATION TESTING Bronchoprovocation testing is typically used to determine the presence or absence of hyperreactive airways disease. Some individuals in whom there is a clinical suspicion of asthma may have normal expiratory flow rates and lung volumes. Bronchoprovocation testing in these individuals can be important in identifying hyperreactive airways disease and supporting the diagnosis of asthma. Methacholine is a cholinergic agonist that causes bronchoconstriction. Individuals with hyperreactive airways exhibit airflow limitation after inhaling low concentrations of methacholine. During the bronchoprovocation test, the subject inhales increasing concentrations of methacholine. Measurements of FEV1 and FVC, as well as specific airways conductance, are obtained after the inhalation of each concentration. If the FEV1 is reduced by 20% or more or the specific airways conductance is reduced by 40% or more, a diagnosis of hyperreactive airways disease can be ascertained. Patients with asthma demonstrate a fall in FEV1 of 20% from baseline at doses considerably smaller than normal individuals (Fig. 16-19).

LUNG DIFFUSION CAPACITY The diffusion of oxygen from the alveolus into the capillary can be assessed by measuring the diffusion capacity for carbon monoxide. To calculate diffusion capacity for oxygen, one needs to know the alveolar volume and the partial pressure of oxygen in the alveolus and in the pulmonary capil-

120

FEV1 (% control)

100

PC20 = 3.0 PC20 = 25

Baseline Saline 0.1 1.0 10.0 100 Methacholine/histamine concentration (mg/mL) Figure 16-19 Bronchoprovocation challenge. Patients are exposed to increasing concentrations of an inhaled challenge (e.g., methacholine, histamine) followed by evaluation of FEV1 (percent control, PC). The FEV1 falls at lower concentrations of the challenge drug in a patient with asthma (blue circles) when compared with an individual without asthma (red circles).

209

lary. Because it is not practical to measure the oxygen tension of pulmonary capillary blood, carbon monoxide is used rather than oxygen to assess diffusion capacity. Carbon monoxide diffuses across the alveolar capillary membranes much as oxygen does. However, carbon monoxide has the advantage of completely binding to hemoglobin. Therefore, the partial pressure of carbon monoxide in the pulmonary venous blood is negligible. Diffusion capacity for carbon monoxide (Dlco) is then measured as the rate of disappearance of carbon monoxide from the alveolus and is used as a surrogate for oxygen diffusion capacity. This measurement provides an overall assessment of gas exchange and depends on factors including the surface area of the lung, the physical properties of the gas, perfusion of ventilated areas, hemoglobin concentration, and the thickness of the alveolar-capillary membrane. Thus, an abnormal Dlco may not only signify disruption of the alveolar-capillary membrane but may also be related to a reduction in surface area of the lung (pneumonectomy), poor perfusion (pulmonary embolus), or poor ventilation of alveolar units. An increased Dlco may be associated with engorgement of the pulmonary circulation with red blood cells or polycythemia. A low Dlco may be seen in interstitial lung diseases that alter the alveolar-capillary membrane or diseases such as emphysema that destroy both alveolar septa and capillaries (Web Fig. 16-4). Anemia lowers the Dlco. Most laboratories provide a hemoglobin correction for diffusion capacity.

ARTERIAL BLOOD GASES The measurement of Pao2 and Paco2 provides information about the adequacy of oxygenation and ventilation. This requires arterial blood sampling through arterial puncture or indwelling cannula (Table 16-2). Oxygenation can also be measured through noninvasive devices including the pulse oximeter, which measures hemoglobin oxygen saturation, and through transcutaneous devices that measure Pao2 and Paco2. These devices are particularly useful for measuring oxygenation during exertion in the office setting. Often, alterations in oxygenation are not detected at rest, but they are unveiled during exertion. The 6-minute walk test is a standardized test in which the patient walks for 6 minutes while the oxygen hemoglobin saturation is measured. A decrease in the oxygen hemoglobin saturation is abnormal and suggests impaired gas exchange capabilities. In summary, pulmonary function tests, in conjunction with history and physical examination, can be used to diagnose pulmonary disorders and assess severity and response to therapy, as illustrated in the flow diagram (Web Fig. 16-5).

Table 16-2 Normal Values for Arterial Blood Gases Po2: 104 − (0.27 × age) Pco2: 36-44 pH: 7.35-7.45 Alveolar-arterial O2 difference = 2.5 + 0.21 × age

210

Section IV—Pulmonary and Critical Care Medicine

Evaluation of Lung Structure CHEST RADIOGRAPH Generally, the evaluation of a patient with lung disease begins with routine chest radiography and then proceeds to more specialized techniques such as computed tomography (CT) or magnetic resonance imaging (MRI). Ideally, the chest radiograph consists of two different films, a postero anterior (PA) radiograph and a lateral radiograph (Web Fig. 16-6). Many pathologic processes can be identified on a PA chest radiograph; however, the lateral view adds valuable information about areas that are not well seen on the PA projection. In particular, the retrocardiac region, the posterior bases of the lung, and the bony structure of the thorax such as the vertebral column are visualized better on the lateral radiograph. The PA chest radiograph is obtained with the patient standing with his or her back to the x-ray beam and the anterior chest wall placed against the film cassette. The chest radiograph should be obtained while the patient takes the deepest breath possible. When the patient is too weak to stand or too sick to travel to the radiology department, the cassette is placed behind the patient’s back, and the x-ray beam travels from anterior to posterior (AP film). The quality of a portable film is not that of a standard PA film, but it still provides valuable information. The approach to examining a chest radiograph should be systematic so that subtle abnormalities are not missed. An examination of a chest radiograph includes evaluating the lungs and pulmonary vasculature, the bony thorax, the heart and great vessels, the diaphragm and pleura, the mediastinum, the soft tissues, and the subdiaphragmatic areas. Abnormalities seen on a chest radiograph include pulmonary infiltrates, nodules, interstitial disease, vascular disease, masses, pleural effusions and thickening, cavitary lung disease, cardiac enlargement, some airway diseases, and vertebral or rib fractures. In addition to the PA and lateral chest radiographs, the lateral decubitus projection is often used to identify the presence or absence of pleural effusion. The decubitus view is particularly useful in determining whether blunting of the costal phrenic sulcus is due to freely flowing pleural fluid or related to pleural thickening. The chest radiograph in concert with a good history and physical examination allows the clinician to diagnose chest disease in many circumstances.

FLUOROSCOPY Fluoroscopic examination of the chest is useful in evaluating motion of the diaphragm. This technique is particularly helpful in diagnosing unilateral diaphragm paralysis. A paralyzed hemidiaphragm moves paradoxically when the patient is instructed to inhale or forcefully sniff. However, fluoroscopy is limited when evaluating for bilateral diaphragm paralysis. Because of compensatory respiratory strategies in the setting of bilateral diaphragm paralysis, apparent normal descent of the diaphragm may be seen during inspiration, leading to a false-negative result by fluoroscopy. Furthermore, paradoxical hemidiaphragm motion is seen in as many as 6% of normal subjects during the sniff maneuver. This observation leads to a false-positive interpretation.

Alternatively, two-dimensional B-mode ultrasound of the diaphragm can be used to visualize diaphragm contraction during inspiration. With this technique, the diaphragm muscle is visualized in the zone of apposition of the diaphragm to the rib cage. Absence of contraction correlates with the absence of active transdiaphragmatic pressure and indicates diaphragm paralysis. This technique can be used to diagnose bilateral and unilateral diaphragm paralysis.

COMPUTED TOMOGRAPHY CT has many applications in pulmonary medicine and provides more detailed information about lung structure than chest radiography. Using this technique, cross sections of the entire thorax can be obtained, usually at 1-cm intervals. Image contrast can be adjusted to optimize visualization of the lung parenchyma or pleural and mediastinal structures. The use of intravenous contrast material as part of the examination permits separation of vascular from nonvascular mediastinal structures. CT of the chest adds tremendous anatomic detail when compared with chest radiography. Its increased resolution permits elucidation of many findings. It helps to characterize pulmonary nodules and masses, distinguish between pleural thickening and pleural fluid, estimate the size of the heart and presence of pericardial fluid, identify patterns of involvement of interstitial lung disease, detect cavities, identify intracavitary processes such as mycetomas, quantify the extent and distribution of emphysema, detect and measure mediastinal adenopathy for staging of lung cancer, and identify vascular invasion by neoplasm. Newer generations of CT scanners are able to use multiple x-ray beams so that 4 to 64 images are created simultaneously at a much faster rate than the older models, which used only a single x-ray beam and detector. CT angiography allows for construction of threedimensional images of the pulmonary vascular system. This imaging technique has emerged as the procedure of choice for identifying pulmonary embolism supplanting pulmonary ventilation-perfusion scintigraphic lung scanning. The technique also can be used to identify pulmonary vascular abnormalities such as aortic dissection, pulmonary venous malformations, and aortic aneurism. High-resolution CT is a technique that generates thin anatomic slices (1mm) to provide a high-contrast image of the pulmonary parenchyma. With high-resolution CT, a special reconstruction algorithm sharpens the soft tissue interfaces to provide superior visualization of the pulmonary parenchyma. This technique primarily is used to identify inter stitial lung disease and bronchiectasis. It is extremely useful in identifying interstitial lung disease that may not be apparent on a plain chest radiograph and has supplanted bronchography in the diagnosis of bronchiectasis.

MAGNETIC RESONANCE IMAGING MRI is a tomographic technique that uses radio waves modified by a strong magnetic field to produce an image. It provides images that are similar to those produced with CT but with better definition of vascular structures. MRIs can be constructed in one of several anatomic planes. Although the standard image is usually an axial view, sagittal and coronal images can be easily created from the information obtained

Chapter 16—Evaluating Lung Structure and Function at the time of the study. Intravenous administration of gadolinium acts as a contrast agent and allows better visualization of vascular structures. MRI can be used to study aortic dissection and may have a role in the evaluation of pulmonary emboli.

PULMONARY ANGIOGRAPHY Pulmonary angiography entails placement of a catheter in the pulmonary artery followed by rapid injection of contrast. Angiography was the gold-standard for diagnosing pulmonary thromboembolic disease. Pulmonary angiography can be useful in detecting congenital abnormalities of the pulmonary vascular tree, but CT and MRI have largely supplanted pulmonary angiography.

POSITRON-EMISSION TOMOGRAPHY Positron-emission tomography (PET) detects metabolically active masses greater than 1cm in diameter. It is helpful in assessing whether a pulmonary nodule is benign or malignant. However, it does not distinguish between inflam mation and malignancy. Thus, assessment of multiple pulmonary nodules using PET scanning is limited because of false-positive findings due to active granulomatous disease such as tuberculosis, sarcoidosis, or fungal infections. Dualmodality integrated PET-CT combines morphologic and functional imaging. The combination of PET and CT is helpful in localizing solitary metastatic lymph nodes in the hilum, which allows better staging of lung cancer. In addition, PET-CT is helpful in planning radiation therapy for patients with lung cancer associated with atelectasis.

BRONCHOSCOPY Fiberoptic bronchoscopy is used for diagnostic or therapeutic indications. It is most commonly performed to directly visualize the nasopharynx, larynx, vocal cords, and proximal tracheobronchial tree for diagnostic purposes. The procedure is performed by sedating the patient and providing

211

local anesthesia with inhaled and bronchoscopically instilled lidocaine. The bronchial mucosa is assessed for endobronchial masses, mucosal integrity, extrinsic compression, dynamic compression, and hemorrhage. The bronchoscope is equipped with a channel for passage of biopsy forceps, bronchial brushes, or needles for aspiration and tissue biopsy. Saline also can be instilled through the channel for bronchial washings or bronchoalveolar lavage. Bronchial washings can be analyzed for cytology, culture, and special stains. A bronchial brush is used to scrape the bronchial mucosa and harvest cells for cytology. Bronchoscopes can also be adapted to provide ultrasound images of the airways and neighboring tissues. Endobronchial ultrasound (EBUS) uses high acoustic frequencies, in the range of 20 MHz, which provide high-resolution images of proximal tissue. EBUS can provide guidance for needle aspiration of mediastinal lymph nodes. Common therapeutic indications for bronchoscopy include the retrieval of foreign bodies, suctioning of secretions, re-expansion of atelectatic lung, and assistance with difficult endotracheal intubations. In special centers, bronchoscopy is used to perform YAG laser therapy of endobronchial lesions, guide placements of catheters for brachytherapy of lung cancer, or guide placement of stents. Lasers produce a beam of light that can induce tissue vaporization, coagulation, and necrosis. Cryotherapy probes induce tissue necrosis through hypothermic cellular crystallization and microthrombosis. Cryotherapy and electrocautery have been used to treat and relieve airway obstruction caused by benign tracheal bronchial tumors, polyps, and granulation tissue. The goal of endobronchial brachytherapy is to relieve airway obstruction from central tumors. This is generally used as an adjunct to conventional external-beam irradiation. Tracheobronchial stenting can be performed to manage airway compression associated with malignant tumors, tracheoesophageal fistulas, and tracheobronchomalacia. Bronchoscopy is generally a safe procedure, with major complications, including significant bleeding, pneumo thorax, and respiratory failure, occurring in 0.1% to 1.7% of patients.

Prospectus for the Future Continued refinement and evolution of techniques and methods currently used to assess pulmonary structure and function will enhance our ability to diagnose and treat individuals with lung disease. Although pulmonary function testing has been performed for decades, advances in equipment design and better standardization of methods will improve accuracy and reproducibility. Further development of noninvasive techniques used to measure changes in lung volume from body surface displacements may allow for assessment of pulmonary function in settings outside of the pulmonary function laboratory. Great strides in assessing lung structure will evolve from advances in CT, PET, and MRI technology. CT volume-rendering techniques will provide images of the central airways enabling “virtual bronchoscopy.” This technique may be useful to guide biopsy location for conventional bronchoscopy and allow visualization of airways distal to an endobronchial obstruction. Volumetric measurements of pulmonary nodules using CT seg-

mentation techniques will allow more accurate calculation of nodule volume and better assessment of tumor doubling times. This, in concert with PET-CT, may provide more accurate means of determining malignant potential of solitary pulmonary nodules. MRI may evolve into the preferred method for evaluating pulmonary emboli and mediastinal disease. Velocityencoded MRI is a promising modality for assessing pulmonary vascular blood flow and pressures, which may prove to be more accurate than current noninvasive methods. Lymph node specific magnetic resonance contrast agents and the development of PET molecular tracers targeting tumor proteins and receptors may better differentiate enlarged lymph nodes due to hyperplasia from those due to neoplasia. Finally, new insights into function of the respiratory control centers in the cortex and brainstem may be attained from studies using functional MRI of the brain.

212

Section IV—Pulmonary and Critical Care Medicine

References McCool FD, Hoppin FG Jr: Respiratory mechanics. In Baum GL (Ed-in-Chief): Textbook of Pulmonary Diseases. Philadelphia, Lippincott-Raven Publishers, 1998, pp 117-130. Miller WT: Radiographic evaluation of the chest. In Fishman AP (Ed-in-Chief): Fishman’s Pulmonary Diseases and Disorders. New York, McGraw-Hill, 2007, pp 455-510.

Wagner PD: Ventilation, pulmonary blood flow, and ventilation-perfusion relationships. In Fishman AP (Ed-in-Chief): Fishman’s Pulmonary Diseases and Disorders. New York, McGraw-Hill, 2007, pp 147-160. West JB: Respiratory Physiology: The Essentials, 5th ed. Baltimore, Williams & Wilkins, 1995. West JB, Wagner PD: Pulmonary gas exchange. Am J Respir Crit Care Med 157:S82-S87, 1988.

Chapter

Obstructive Lung Diseases Matthew D. Jankowich

he obstructive lung diseases are a group of common pulmonary disorders resulting in dyspnea charac terized by an obstructive pattern of expiratory airflow limitation on spirometry. These disorders include chronic obstructive pulmonary disease (COPD), asthma, cystic fibrosis, bronchiectasis, and the bronchiolar disorders. COPD is a clinical term encompassing varied pathophysiologic processes, including emphysema, chronic bronchitis, and small airways disease, one or more of which may be prominent in a given patient with this disorder. COPD is characterized in general by abnormal airway inflammation and abnormal lung structure in response to an inhaled irritant, typically cigarette smoke, resulting in irreversible or incompletely reversible airflow limitation. Asthma is distinguished from COPD by characteristic bronchial smooth muscle hyperreactivity and reversible airflow limitation and by its frequent association with atopy (Fig. 17-1). These disorders are epidemic in the general population and account for a significant proportion of the morbidity and mortality associated with the obstructive lung diseases. All the obstructive lung disorders cause an obstructive pattern of expiratory airflow limitation, although the basis for airflow obstruction varies among disorders. The flow of air through the bronchial tree is directly proportional to the driving pressure and is inversely proportional to the resistance. In obstructive lung disease, alterations in one or both of these processes might be present. For example, in emphysema, airflow limitation is caused by decreased elastic recoil resulting in decreased driving pressure. By contrast, in asthma, airflow limitation is due to bronchoconstriction that increases airway resistance. Airway obstruction to flow causes characteristic changes in lung volumes. The residual volume and functional residual capacity are increased, whereas the total lung capacity remains normal or is increased. Vital capacity is reduced by the increase in residual volume. Several factors may contribute to the increase in functional residual capacity and residual volume in obstructive lung disease. Decreased lung elastic recoil in emphysema increases the functional residual capacity because of reduced opposition to the outward force exerted by the chest wall. Loss of airway tone and decreased tethering by surrounding lung in COPD, as well as bronchoconstriction and mucus plugging in acute asthma, allow airways to collapse at higher

lung volumes and trap excessive air. Finally, under demands for increased minute ventilation such as during exercise, the increased resistance to airflow may not allow the lungs to empty completely during the time available for expiration, leading to so-called dynamic hyperinflation of the lungs as the volume of trapped air progressively increases while the inspiratory capacity is progressively limited. The three major consequences of the changes in lung volume seen with obstructive lung disease are as follows: (1) Breathing at higher lung volumes requires a higher change in pressure for the same change in lung volume, and this requirement increases the work of breathing. (2) Larger lung volumes place the inspiratory muscles at a mechanical disadvantage. The diaphragm is flattened, thereby decreasing its ability to change intrathoracic volume, and all the inspiratory muscle fibers are shortened, decreasing the tension they are able to exert to effect changes in lung volume. (3) Lung volumes are larger, resulting in tethering of the narrowed and collapsing airways by the surrounding lung parenchyma, tending to retain airway patency and reduce airway resistance and air trapping; this consequence is beneficial. These three physiologic derangements explain many of the clinical features of obstructive lung diseases (Table 17-1). Although the consequences to lung function are relatively similar in the obstructive lung diseases, their pathogenesis, treatment, and prognosis are different. Therefore, a careful evaluation is needed to reach a definitive diagnosis that will guide targeted therapy.

Chronic Obstructive Pulmonary Disease COPD results in slowly progressive dyspnea and is characterized by abnormalities of airway and lung structure occurring in response to noxious inhaled substances, especially cigarette smoke. The abnormalities of airway and lung structure in COPD result in irreversible airflow limitation, the physiologic hallmark of COPD. The term COPD encompasses emphysema, chronic bronchitis, and small airways disease, pulmonary disorders that have common clinical 213

214

Section IV—Pulmonary and Critical Care Medicine

manifestations and often co-exist in the same patient. The term excludes other causes of airflow obstruction, such as asthma, although in practice these diseases may at times overlap. COPD is one of the most common disorders seen by physicians, and it is the fourth leading cause of death in the United States. There are an estimated 1,500,000 emergency department visits related to COPD annually, and about 700,000 patients with COPD are hospitalized each year. From 1997 to 2001, about 90,000 deaths per year attributable to COPD occurred in the United States. Although COPD has historically been more prevalent in males than females, the prevalence of COPD in females has been increasing, and annual death rates for COPD have been steadily rising in

Chronic bronchitis

Emphysema Other disorders with airflow obstruction Chronic bronchiolitis

COPD

ABPA Other causes Asthma Cystic fibrosis Bronchiectasis Figure 17-1 Classification of obstructive lung diseases. COPD, chronic obstructive lung disease.

both white and black women. Prevalence rates for COPD are correlated with increasing age, lower socioeconomic status, and smoking. Cigarette smoking is by far the most common cause of COPD; however, other factors such as inhalation of cooking fire smoke, air pollution, occupational exposures to dust and fumes, and infections contribute to the occurrence, severity, and progression of the disease. Although cigarette smoking is the most common cause, it is important to note that only 20% of smokers are thought to develop clinically significant COPD (although many more may experience some loss of lung function). This finding suggests that COPD results from a susceptibility to environmental factors (e.g., tobacco) as a result of a genetic predisposition. A genetic predisposition is also implied by the documentation of familial clusters of COPD. Several longitudinal studies have defined patterns of age-related decline in lung function and have documented the concept of susceptibility to COPD. These studies show that most adult nonsmoking men exhibit a decline in forced expiratory volume in 1 second (FEV1) of 35 to 40 mL per year. This rate is increased to 45 to 60mL per year in most cigarette smokers. However, the susceptible smoker may demonstrate losses of 70 to 120mL per year (Fig. 17-2). This information allows the physician to project the rate of decrease of lung function in patients with COPD and assess the effects of therapeutic interventions. Although COPD results in chronic, progressive dyspnea, periodic acute exacerbations are characteristic of COPD. Exacerbations are characterized by a rapidly developing worsening of pulmonary function respiratory symptoms such as cough and sputum production. Acute exacerbations are associated with various triggers, including viral or bacterial respiratory infections, air pollution, and cardiac failure. Exacerbations vary widely in severity, but severe exacerbations may lead to hospitalization, acute respiratory failure, and death. Following an exacerbation, a patient may take weeks to return to a baseline level of function. Patients with

Table 17-1 Features of Obstructive Lung Diseases Disorder

Clinical Features

Laboratory Findings

Chronic obstructive pulmonary disease Emphysema

Chronic progressive dyspnea

Decreased expiratory flow rates, hypoxia and hypercapnia in end-stage disease Hyperinflation, increased compliance, low DLco, rarely α1-antitrypsin deficiency Nonspecific; rarely occurs in isolation without varying degrees of emphysema Airway hyperreactivity, response to bronchodilators

Chronic bronchitis Asthma Bronchiectasis Immotile cilia syndrome Hypogammaglobulinemia Cystic fibrosis

Little or no sputum, end-stage cachexia Sputum, history of smoking, industrial exposure Episodic dyspnea, cough, wheezing, with or without environmental triggers Usually large volume of sputum Situs inversus, dextrocardia, sinusitis, infertility Sinusitis, bronchiectasis, meconium ileus, malabsorption, infertility

Chest radiograph: dilated bronchi, thick-walled, tram track shadows, obstruction with or without restriction on pulmonary function tests Abnormal dynein in ciliated cells Decrease in one or more immunoglobulins Increased sweat chloride, mutation in CFTR chloride channel, elevated fecal fat, abnormal nasal mucosal potential difference

CFTR, cystic fibrosis transmembrane conductance regulator; DLco, diffusion capacity for carbon monoxide.

Chapter 17—Obstructive Lung Diseases FEV1 (% of value at age 25)

100

Never smoked or not susceptible to smoke

75 50

Smoked regularly and susceptible to its effects

Stopped at 45

Disability 25

Stopped at 65

Death 0

75 Age (yr) Figure 17-2 Pattern of decline in forced expiratory volume in 1 second (FEV1) with risks for morbidity and mortality from respiration disease in a susceptible smoker in comparison with a normal patient and with a nonsusceptible smoker. Although cessation of smoking does not replenish the lung function already lost in a susceptible smoker, it decreases the rate of further decline. (Data from Fletcher C, Peto R: The natural history of chronic airflow obstruction. BMJ 1:1645-1648, 1977.) 25

frequent exacerbations of COPD appear to experience an accelerated rate of decline in FEV1. The only genetic disorder thus far definitively linked to COPD is α1-antitrypsin deficiency, which accounts for less than 1% of all cases. The deficient enzyme, α1-antitrypsin, an acute-phase reactant, is produced primarily in the liver, from which it travels to the lung, where it deactivates elastases released by inflammatory cells that are capable of degrading connective tissue matrices. In doing so, α1-antitrypsin prevents the uncontrolled degradation of elastin in the lung parenchyma and protects against the development of emphysema. Individuals with the ZZ genotype of α1antitrypsin deficiency produce mutant forms of α1antitrypsin that have a tendency to inappropriately polymerize within the hepatocyte, leading to a deficiency in secreted α1-antitrypsin and in some cases to collateral damage to the liver caused by accumulation of intracellular misfolded, mutant α1-antitrypsin. Patients who develop emphysema at a young age (400mg/day Acute Weeks or months Subacute or chronic

Pneumonitis, fibrosis

Acute

Pulmonary edema, bronchospasm

Opiates Cocaine

Acute Acute

Talc (in intravenous and inhaled illicit drugs)

Acute or chronic

Pulmonary edema Pulmonary edema, diffuse alveolar damage, pulmonary hemorrhage, BOOP Granulomatous interstitial fibrosis, granulomatous pulmonary artery occlusion, particulate embolization

Chemotherapeutic Bevacizumab Bleomycin

Antimicrobial Nitrofurantoin Sulfasalazine Cardiovascular Amiodarone Flecainide Tocainide Procainamide

ARDS, LIP Pneumonitis Drug-induced systemic lupus erythematosus, pleural effusions, pulmonary infiltrates

Anti-inflammatory Aspirin Illicit

Tocolytics Terbutaline, albuterol, ritodrine

Acute

Pulmonary edema

ARDS, acute respiratory distress syndrome; BOOP, bronchiolitis obliterans and organizing pneumonia; LIP, lymphoid interstitial pneumonia.

high levels of inspired oxygen may precipitate bleomycin lung injury and should be avoided if possible in exposed patients. The website http://www.pneumotox.com is an online reference site tabulating the reported pulmonary toxicities of various drugs that is searchable by drug name as well as by pattern of lung involvement.

Pulmonary Vasculitis and Diffuse Alveolar Hemorrhage Diffuse alveolar hemorrhage (DAH) syndromes encompass diverse group of specific entities that are all characterized by the disruption of the alveolar-capillary membrane, resulting in bleeding into the alveolar spaces. Unfortunately, patients with DAH do not always present with signs, symptoms, and

laboratory and radiographic findings that support a specific underlying diagnosis. Often, DAH is found without features that identify a specific etiology. All DAH syndromes are characterized by the abrupt onset of cough, fever, and dyspnea. Hemoptysis is common but not universal because it may be absent in up to one third of patients with DAH. Physical findings are generally nonspecific, although ocular, nasopharyngeal, or cutaneous abnormalities may suggest systemic vasculitis or collagen vascular disease as an etiology. The cardiopulmonary examination is often normal but may reveal inspiratory crackles, a systolic murmur suggestive of mitral stenosis or evidence of pulmonary hypertension. Falling hemoglobin levels, the presence of increasingly hemorrhagic fluid on sequential bronchoalveolar lavage and new patchy alveolar infiltrates (Web Fig 18-6) by chest imaging favor the diagnosis of DAH. Other laboratory abnormalities may include the presence of azotemia, suggesting a pulmonary-renal syndrome. In this setting, an abnormal urinalysis with proteinuria, hematuria, and red

236

Section IV—Pulmonary and Critical Care Medicine

blood cell casts is usually seen. The erythrocyte sedimentation rate (ESR) may be increased, particularly in those with an underlying systemic disease. Some lung disorders characterized by DAH are associated with the production of antineutrophil cytoplasmic antibodies (ANCA) directed against neutrophil cytoplasmic antigens or antibodies directed at the glomerular basement membrane. ANCA testing, in particular, can play an important role in the workup of DAH because it is used in the diagnosis and classification of various pulmonary vasculitides that cause DAH. Two major immunofluorescent patterns can be seen in ANCA testing: diffuse staining throughout the cytoplasm (c-ANCA) or staining around the nucleus (p-ANCA). Specific antigens that ANCAs are directed against include proteinase-3 (PR-3), typically causing the c-ANCA pattern, and myeloperoxidase (MPO), which typically causes the p-ANCA pattern. All DAH syndromes are characterized by three distinct histologic patterns. Bland pulmonary hemorrhage is due to alveolar hemorrhage without inflammation or destruction of the alveolar structures. This pattern is seen in conditions where there is elevated pulmonary capillary hydrostatic pressure, such as congestive heart failure or mitral stenosis, or with the use of certain anticoagulation medications. Diffuse alveolar damage (DAD) is caused by a variety of pulmonary infections, connective tissue diseases, and medications. DAD is also seen in acute respiratory distress syndrome (ARDS) from any etiology. Histologically, alveolar walls appear edematous and are lined with hyaline membranes. The most common histologic pattern seen on lung biopsy obtained from patients with DAH is pulmonary capillaritis, which is characterized by neutrophilic infiltration of the alveolar septa. This sequentially leads to necrosis, loss of capillary structural integrity, and extravasation of red blood cells into the interstitium and alveolar spaces. This finding is seen in a variety of connective tissue diseases and is the most common histologic pattern of the pulmonary vasculitides. The pulmonary vasculitides represent a group of specific entities, many of which are associated with elevated serum ANCA levels. These entities include Wegener granulomatosis, microscopic polyangiitis, Churg-Strauss syndrome, and certain drug-induced vasculitis syndromes. Pauci-immune glomerulonephritis without evidence of extrarenal disease is a disorder that is considered to be on the spectrum of Wegener granulomatosis and microscopic polyangiitis because its histologic features are indistinguishable from these disorders and some patients eventually develop extrarenal (pulmonary) manifestations. Wegener granulomatosis is a systemic necrotizing granulomatous vasculitis that often involves the small and medium-sized vessels of the upper airway, the lower respiratory tract, and the kidney. Although this triad is not always seen at initial presentation, as only 40% of those affected have renal disease at that time, 80% to 90% of patients eventually develop glomerulonephritis. The most frequent manifestations of this illness are pulmonary, as highlighted by cough, chest pain, hemoptysis, and dyspnea. Constitutional symptoms, such as fever and weight loss, as well as symptoms due to involvement of the skin, eye, heart, nervous system, and musculoskeletal system, are also common. The diagnosis of Wegener granulomatosis is supported by clinical findings and by the presence of circulating ANCAs, which are seen in 90% of all patients. The remaining 10% are ANCA negative. In ANCA-positive patients, antibodies are

usually directed against PR-3; however, 10% to 20% may have anti-MPO antibodies. Chest imaging may show bilateral disease and infiltrates that evolve over the course of the illness. Lung nodules are common and may cavitate. Effusions and adenopathy are not common. Sinus films or CT scans serve to diagnose upper airway involvement. Tissue biopsy at a site of active disease is generally needed to confirm Wegener’s granulomatosis. The presence of granulomatous inflammation is common, but actual vasculitis is seen in only 35% of patients. A renal biopsy is preferred because it is easier to perform and more often diagnostic. In the absence of renal involvement, a lung biopsy should be considered. Pathologically, Wegener granulomatosis is characterized by small and medium vessel necrotizing vasculitis and granulomatous inflammation. Special stains and cultures should be performed to exclude the presence of infections that can produce similar findings. Microscopic polyangiitis is a form of systemic necrotizing small vessel vasculitis that universally affects the kidneys, whereas pulmonary involvement occurs in only a minority of patients (10% to 30%). This rare condition has a prevalence of 1 to 3 cases per 100,000, but it is the most common cause of pulmonary-renal syndrome. It is often heralded by a long prodromal phase, characterized by constitutional symptoms followed by the development of rapidly progressive glomerulonephritis (RPGN). In those patients who do develop lung involvement, DAH secondary to capillaritis is the most common manifestation. Joint, skin, peripheral nervous system, and gastrointestinal involvement can be seen as well. Seventy percent of patients with microscopic polyangiitis are ANCA positive, most of whom have antiMPO antibodies. Because anti-MPO and anti-PR3 antibodies can be present in both microscopic polyangiitis and Wegener granulomatosis, these diseases cannot be distinguished based on their ANCA pattern. However, the two diseases can be distinguished pathologically because microscopic polyangiitis is characterized by a focal, segmental necrotizing vasculitis affecting venules, capillaries, arterioles, and small arteries without clinical or pathologic evidence of necrotizing granulomatous inflammation. The absence or paucity of immunoglobulin localization in vessel walls distinguishes microscopic polyangiitis from immune complex– mediated small vessel vasculitis such as Henoch-Schönlein purpura and cryoglobulinemic vasculitis. Treatments of Wegener granulomatosis and microscopic polyangiitis are similar. Combination therapy with cortico steroids and cyclophosphamide is the standard of care. Azathioprine can be substituted for cyclophosphamide if remission is achieved. Intravenous immunoglobulin may be effective for those with persistent disease. Novel therapies, including trimethoprim-sulfamethoxazole, antilymphocyte monoclonal antibodies, and tumor necrosis factor inhibitors, have been tried with some success. Allergic granulomatosis or Churg-Strauss syndrome is characterized by the triad of asthma, hypereosinophilia, and necrotizing vasculitis. Many other organ systems, including the nervous system, skin, heart, and gastrointestinal tract, may be involved as well. The vasculitis can be associated with skin nodules and purpura. Although DAH and glomerulonephritis may occur, they are much less common than in the other small vessel vasculitides. Morbidity and mortality are often due to cardiac or gastrointestinal complications or

Chapter 18—Interstitial Lung Diseases to status asthmaticus and respiratory failure. ANCAs are less helpful in Churg-Strauss syndrome because only 50% of patients are ANCA positive. Anti-MPO antibodies are more commonly seen in these patients. Pathologically, both a necrotizing, small vessel vasculitis and an eosinophil-rich inflammatory infiltrate with necrotizing granulomas are seen. Most patients respond well to corticosteroids, but other immunosuppressants similar to cyclophosphamide may be required in patients with refractory disorders. Other well-known causes of pulmonary capillaritis include the systemic vasculitides, collagen vascular disorders, antiglomerular membrane antibody syndrome (Goodpasture syndrome), and Henoch-Schönlein purpura. Goodpasture syndrome causes DAH associated with glomerulonephritis caused by antiglomerular basement membrane antibodies to the α3 chain of type IV collagen that is also found in the lung basement membrane. More than 90% of patients with Goodpasture syndrome have antiglomerular basement membrane antibodies detectable in the serum. In those without circulating antibodies, the diagnosis may be confirmed by lung biopsy, although the kidney is the preferred site. Up to 40% may also be ANCA positive, primarily with anti-MPO antibodies. Pathologically, linear deposition of antibody along the alveolar or glomerular basement membrane that is visible by direct immunofluorescence occurs. The treatment of Goodpasture syndrome is plasmapheresis and immunosuppression. The disease is fatal if left untreated. Idiopathic pulmonary hemorrhage or hemosiderosis is a diagnosis of exclusion. Patients with this syndrome have recurrent DAH without associated renal or systemic disease. Histologically, the lung shows hemorrhage and hemosiderin accumulation without inflammation. Treatment includes supportive care, immunosuppression, and occasionally plasmapheresis, but response to therapy is varied. This syndrome is most common in children, who have a worse prognosis than adults.

Environmental and Occupational Interstitial Lung Diseases Several environmental and occupational exposures may cause ILDs. These include the pneumoconioses, drug-induced ILD (discussed earlier), and hypersensitivity pneumonitis. Pneumoconiosis and hypersensitivity pneumonitis are discussed later. The pneumoconioses are lung diseases resulting from the inhalation of mineral dusts, including silica, coal dust, or asbestos. Hypersensitivity pneumonitis is caused by the inhalation of organic dusts.

PNEUMOCONIOSIS The pneumoconioses result from the effects of accumulation of mineral dusts in the lungs, with the typical reaction being fibrosis. In general, the risk and extent of these diseases are related to the intensity and cumulative amount of exposure over time. Prevention of the pneumoconioses through occupational safeguards or, in the case of asbestos, legislative bans on use, is most important because there are not effective treatments for these diseases once established.

237

Silicosis is a lung disease caused by exposure to crystallinefree silica, which results in an inflammatory and fibrotic reaction resulting in the characteristic silicotic nodule. Certain occupations that have a higher propensity for exposure to silica include mining, stone cutting, carving, polishing, foundry work, and abrasive clearing (sandblasting). Although exposure is usually chronic (over years), accelerated and acute disease manifestations have been described in the setting of heavier short-term exposures. Acute silicosis causes a pulmonary alveolar proteinosis, with an accumulation of surfactant in the alveolar spaces. Chronic silicosis results in simple nodular silicosis, which is usually asymptomatic unless the patient is also exposed to tobacco, and progressive massive fibrosis (PMF), which is characterized by extensive bilateral apical fibrosis resulting from the confluence of many silicotic nodules. Patients with silicosis may present with dyspnea or may be relatively asymptomatic but present for evaluation of an abnormal chest radiograph. Chest radiographs in uncomplicated silicosis show upper lobe nodular opacities, which may be subtle, whereas PMF results in marked architectural distortion of the upper lobes (Web Fig 18-7). Hilar node enlargement may be seen accompanied by eggshell nodal calcification (Web Fig 18-8). Pulmonary function tests in simple nodular silicosis may be normal or show a mixed obstructive or restrictive pattern, whereas PMF is typically associated with severe restriction and hypoxemia. Patients with silicosis are at elevated risk for tuberculosis and should be screened for latent tuberculosis infection; there is also an association between silicosis and rheumatoid arthritis. Coal worker’s pneumoconiosis is an uncommon cause of pulmonary fibrosis, occurring in workers exposed to coal dust and graphite. Usually, the patients are exposed while working in underground mines. Coal worker’s pneumoconiosis results in the formation of pigmented lesions in the lung, surrounded by emphysema, known as coal macules. PMF may subsequently occur. Most patients show chronic cough, which is usually productive, because of bronchitis related to coal exposure or to tobacco. The chest radiograph shows diffuse small rounded opacities. As with silicosis, there is an association with rheumatoid arthritis; Caplan syndrome is the occurrence of multiple large, sometimes cavitary, lung nodules in association with rheumatoid arthritis following coal dust exposure. Asbestosis is due to chronic exposure to asbestos, which is a fibrous silicate used for insulation, for friction-bearing surfaces, and to strengthen materials. The inhaled asbestos fibers are deposited in the lungs, where small fibers may be phagocytosed and cleared through lymphatics to the pleural space, but longer fibers are often retained. Typically, asbestos exposure may lead to pleural disease characterized by pleural plaques, effusion, and fibrosis, but it does not necessarily affect the lung parenchyma. If it does, it is called asbestosis and is associated with interstitial lung fibrosis. Asbestosis is characterized by a gradual onset of dyspnea. As with other pneumoconioses, the risk and severity of disease are related to the extent and duration of cumulative exposure. Asbestosis is often diagnosed after exposure has ceased, and disease progression may continue to occur in the absence of ongoing exposure, owing to the reaction to retained asbestos fibers in the lung. The clinical presentation, pulmonary function tests, and imaging studies are similar to those

238

Section IV—Pulmonary and Critical Care Medicine

found in restrictive lung diseases like IPF. However, the detection of significant pleural disease is useful in distinguishing this illness from other ILDs. The diagnosis is made from the history and demonstration of concomitant pleural plaques and lower lobe–predominant fibrotic changes on chest radiographs or CT scan. In uncertain cases, the demonstration of asbestos in tissue specimens may be necessary; asbestos bodies are the characteristic finding and consist of asbestos fibers coated by iron-containing (ferruginous) material. Asbestos exposure results in an elevated incidence of malignancy, including lung carcinoma and mesothelioma, especially in persons who also smoke. Whether the presence of asbestosis itself confers a heightened risk for malignancy, independent of the effects of asbestos exposure alone, is uncertain. No specific treatment for asbestosis exists. Berylliosis results from exposure to beryllium, a rare metal useful in modern, high-technology industries. Exposure to beryllium can lead to an acute chemical bronchitis and pneumonitis or chronic beryllium disease. Chronic beryllium disease is characterized by a multisystemic granulomatosis that is difficult to distinguish from sarcoidosis. The diagnosis is made by history of exposure, histologic examination, and laboratory confirmation through the beryllium lymphocyte proliferation test that is available at specialized centers. Corticosteroids may be useful in the treatment of berylliosis, but patients should avoid further exposure to beryllium.

HYPERSENSITIVITY PNEUMONITIS Hypersensitivity pneumonitis (also termed extrinsic allergic alveolitis) is a relatively common ILD resulting from an exaggerated immune reaction to various inhaled organic antigens in sensitized individuals. Potential sensitizing antigens are diverse, ranging from bacterial, fungal, and animal proteins to low–molecular-weight chemicals (Table 18-7). Although evocative descriptions have been given to occupational forms of this disease (“paprika splitter’s lung” resulting from sensitivity to Mucor stolonifer), more prosaic exposures may occur in everyday life, for example, to contaminated hot tub water or to pet birds. The disease may present in an acute fashion several hours after intense exposure to a provocative antigen, with fever, chills, cough, dyspnea, and malaise that last for up to 24 hours. Subacute or chronic disease may occur with repeated or prolonged antigen exposure and may result in chronic dyspnea and cough, with eventual progression to pulmonary fibrosis. Diffuse crackles and wheezes are common physical findings. Hypoxemia may be present. In general, hypersensitivity pneumonitis is characterized by nonspecific infiltrates in the mid and upper lung fields on chest radiographs. CT scanning is more sensitive than chest radiography, revealing ground-glass opacities, centrilobular nodules, and mosaic attenuation patterns resulting from airway obstruction. In chronic hypersensitivity pneumonitis, emphysema and lower lobe honeycombing may be present. Restrictive or mixed obstructive-restrictive patterns may be seen on pulmonary function testing, along with abnormalities of gas exchange. Bronchoalveolar lavage may demonstrate a lymphocytic alveolitis, with CD8 T-lymphocyte predominance. Patients with hypersensitivity pneumonitis may have pre-

Table 18-7 Hypersensitivity Pneumonitis Antigen

Source

Disease Examples

Thermophilic bacteria

Moldy hay, sugar cane, compost

Other bacteria

Contaminated water, wood dust, fertilizer, paprika dust

Fungi

Moldy cork, contaminated wood dust, barley, maple logs

Animal protein

Bird droppings, animal urine, bovine and porcine pituitary powder

Chemically altered human proteins (albumin and others) Phthalic anhydride

Toluene diisocyanate Trimellitic anhydride Diphenylmethane diisocyanate Heated epoxy resin

Farmer’s lung, bagassosis, mushroom worker’s disease Humidifier, detergent worker’s disease, and familial hypersensitivity pneumonitis Suberosis, sequoiosis, and maple bark stripper’s disease, malt worker’s disease, and paprika splitter’s lung Pigeon breeder’s lung, duck fever, turkey handler’s disease, pituitary snuff taker’s disease, laboratory worker’s hypersensitivity pneumonitis Hypersensitivity pneumonitis

Epoxy resin lung

cipitating antibodies to the offending antigen, but serum precipitins are not sufficiently sensitive or specific for diagnosis, and the specific antigen may not be known or may not be tested for with standard test panels. An appropriate exposure, clinical history, and imaging findings can suggest the diagnosis, but lung biopsy may be necessary for confirmation. Transbronchial biopsy is often sufficient. Typical biopsy findings include poorly formed granulomas containing foreign body giant cells and interstitial chronic inflammation with a bronchiolocentric component. Clinical improvement often occurs in the hospital setting when patients are isolated from the offending antigen, and relapse may occur after discharge. Corticosteroids can relieve symptoms in the acute phase, but their efficacy in chronic forms of the disease is less clear. Identification of the cause of hypersensitivity pneumonitis is important because chronic disease management requires avoidance of exposure to the antigen, which can be financially or psychologically challenging for patients in the setting of occupational, pet, or residential exposures.

Specific Entities PULMONARY LANGERHANS CELL HISTIOCYTOSIS (EOSINOPHILIC GRANULOMA) Pulmonary Langerhans cell histiocytosis (LCH), also called eosinophilic granuloma, is a disease of young- to middle-aged adults. Nearly all cases occur in white men who smoke. The

Chapter 18—Interstitial Lung Diseases disorder results from the infiltration of Langerhans cells, which are dendritic cells, into the lung parenchyma. Smoking may alter local immune signaling, attracting the Langerhans cells to the lungs. Patients typically exhibit constitutional symptoms, dyspnea on exertion, and cough, possibly with hemoptysis. Pneumothorax may also occur. Imaging shows micronodular lesions and cysts that predominate in the mid and upper lung zones. Pulmonary function tests show an obstructive pattern and impaired diffusion capacity. Specific diagnosis can be made with open lung biopsy, which demonstrates multiple stellate lung nodules that may be cellular or fibrotic, containing Langerhans cells that stain for Cd1a and S100. Electron microscopy may reveal Birbeck granules, distinctive racquet-shaped structures within the cells. In the right clinical setting and with a typical HRCT, a biopsy might not be needed for diagnosis. In contrast to systemic LCH, pulmonary LCH is not a neoplastic disorder, and spontaneous regression may occur. The main treatment is tobacco cessation. Corticosteroids and other immunosuppressants are sometimes employed as adjunctive therapy.

LYMPHANGIOLEIOMYOMATOSIS Lymphangioleiomyomatosis (LAM) is a rare disorder that may occur in association with the tuberous sclerosis complex or sporadically in women of childbearing age. The disease is characterized by extensive nodular infiltration of the lungs and lymphatics with growths of smooth muscle–like cells. Mutations in the TSC-1 or TSC-2 gene, encoding for tumor suppressor proteins that normally act as inhibitors of protein synthesis and cell growth, may result in tuberous sclerosis or LAM, with mutations in TSC-2 being associated with greater disease severity. Dyspnea and pneumothorax are the most common presentations, with chylous pleural effusions and hemoptysis also occurring. These clinical presentations result from the lung parenchymal destruction, airway narrowing, and lymphatic obstruction caused by the abnormal proliferation of the smooth muscle–like cells. Imaging studies show an interstitial pattern with mid and upper lung predominance, multiple thin-walled cystic lesions, and characteristically preserved lung volumes. Pleural effusion or pneumothorax may also be present on imaging. CT of the abdomen may reveal fat-containing kidney lesions consistent with angiomyolipomas. Pulmonary function tests typically show a progressive obstructive pattern, although mixed obstruction and restriction may also be seen. Although the clinical features coupled with characteristic imaging are often diagnostic, lung biopsy might be necessary in some cases. This demonstrates interstitial nodules composed centrally of spindle-shaped cells that stain for smooth muscle cell actin as well as with HMB-45, an antibody to the melanocytic glycoprotein gp-100, involving the alveolar walls, lobular septa, venules, small airways, and pleura. Treatment involves management of pleural complications, including use of pleurodesis to prevent recurrent pneumo thorax or effusion; bronchodilator and oxygen therapy; and avoidance of pharmacologic estrogens, which may exacerbate the disease. Progesterones have been used in an attempt to modulate disease progression, although efficacy data are limited. Because the products of the TSC-1 and TSC-2 genes normally act as inhibitors of the mammalian target of rapamycin (m-TOR), use of inhibitors of m-TOR activity

239

like sirolimus is under investigation in LAM, with sirolimus being shown in one small study to improve lung function in LAM. Lung transplantation can be performed in patients with severe pulmonary dysfunction.

EOSINOPHILIC LUNG DISEASE Eosinophilic lung diseases are characterized by the presence of pulmonary infiltrates and eosinophilia of the peripheral blood or lung. Because eosinophilia is a feature of many diseases, it is important to distinguish primary pulmonary eosinophilic lung disorders from lung disorders in which eosinophilia is secondary to a specific cause. Eosinophilic lung diseases can be categorized as follows: primary pulmonary eosinophilic disorders (including acute and chronic eosinophilic pneumonia, hypereosinophilic syndrome), pulmonary disorders of known cause associated with eosinophilia (including asthma, allergic bronchopulmonary aspergillosis, drug reactions, parasitic infections), lung diseases associated with eosinophilia (including hypersensitivity pneumonitis, COP, IPF), malignancies associated with eosinophilia (including lung cancer, leukemia, lymphoma), and systemic disease associated with eosinophilia (including rheumatoid arthritis, sarcoidosis, and Sjögren syndrome). Acute eosinophilic pneumonia is characterized by fever, a nonproductive cough, and dyspnea of less than 7 days’ duration, often leading to respiratory failure. This disease typically affects men between the ages of 20 and 40 years who are otherwise healthy. Chest imaging reveals diffuse bilateral pulmonary infiltrates. Eosinophilia is not present in the peripheral blood initially but may occur 7 to 30 days after onset. However, abundant eosinophils can be found in bronchoalveolar lavage fluid, and a level of greater than 25% of all nucleated cells is helpful in making the correct diagnosis. Although lung biopsy is typically not required to make the diagnosis, it can show eosinophilic infiltration with acute and organizing diffuse alveolar damage. Treatment with corticosteroids typically offers complete clinical and radiographic resolution without recurrence or residual sequelae. Chronic eosinophilic pneumonia is an idiopathic disease predominantly of middle-aged women. Also termed prolonged pulmonary eosinophilia, this illness is characterized by a productive cough, dyspnea, malaise, weight loss, night sweats, and fever associated with progressive peripheral lung infiltrates that, on chest radiography, have been described as resembling the “photographic negative of pulmonary edema” (Web Fig 18-9). On presentation, most patients have a peripheral eosinophilia of greater than 30% and bronchoalveolar lavage fluid eosinophilia as well. Histologic examination shows eosinophils and histiocytes in the lung parenchyma and interstitium, areas of COP, but minimal fibrosis. Spontaneous remissions have been reported, but respiratory failure can also develop. Typically, treatment with corticosteroids is rapidly effective. Prolonged therapy is recommended because unlike with acute eosinophilic pneumonia, relapses are common. Simple pulmonary eosinophilia, also known as Löffler syndrome, is characterized by transient migratory infiltrates that last less than 1 month. In some cases, no symptoms are present, but dyspnea and dry cough may occur. Pathologic examination of tissues reveals interstitial and intra-alveolar

240

Section IV—Pulmonary and Critical Care Medicine

accumulation of eosinophils, macrophages, and edema. The syndrome might be idiopathic or caused by parasitic infections (e.g., Ascaris species, Strongyloides species, hookworms) or drugs (e.g., nitrofurantoin, minocycline, sulfonamides, penicillin, nonsteroidal anti-inflammatory drugs). Treatment requires removal of the offending agent or treatment of the parasitic infection. In idiopathic cases, corticosteroids may be used. Allergic bronchopulmonary aspergillosis (ABPA) is a hypersensitivity reaction that occurs when Aspergillus species colonizes the airways in patients with asthma or cystic fibrosis. Patients may present with fever, malaise, a cough productive of thick brown mucus plugs, and occasionally hemoptysis. On chest radiograph, pulmonary infiltrates, which are often transient and migratory, and central bronchiectasis may be seen. Peripheral eosinophilia of greater than 10%, elevated IgE levels (as well as the presence of aspergillus-specific IgE), and precipitating antibodies to aspergillus are among the laboratory abnormalities seen in ABPA. Response to corticosteroids is good. Itraconazole can be added to the treatment regimen as well.

PULMONARY ALVEOLAR PROTEINOSIS Pulmonary alveolar proteinosis (PAP) is a rare disorder in which lipoproteinaceous material, similar to surfactant, accumulates within the alveoli. PAP has a congenital form,

characterized by mutations of the genes encoding surfactant protein B or C or for the receptor for granulocyte-macrophage colony-stimulating factor (GM-CSF). Secondary PAP occurs in conditions in which there is a functional impairment or decrease in the number of alveolar macrophages, as seen in various hematologic malignancies (leukemia), infections (pneumocystis pneumonia), and inhalation of toxic dusts (silica, aluminum) or following allogeneic bone marrow transplantation. Acquired or idiopathic forms of PAP may represent an autoimmune disease, with neutralizing antibodies directly targeting GM-CSF playing a role in its pathogenesis. Patients with PAP present with progressive dyspnea on exertion, malaise, low-grade fever, and cough. Chest radiograph typically reveals bilateral perihilar opacities. CT scan may show thickening of the intralobular and interlobular septae, creating a pattern referred to as crazy paving, which is a nonspecific finding because it is seen in many other diseases of the lung. Bronchoalveolar lavage fluid can establish the diagnosis because the lavage fluid has a milky, opaque appearance that contains large “foamy” alveolar macrophages with few inflammatory cells. Asymptomatic patients and those with mild symptoms require no immediate treatment. Whole-lung lavage is indicated for patients with hypoxemia or severe dyspnea and, in up to 40% of patients, may be required only one time. GM-CSF administration in patients with acquired PAP may be beneficial, but less so than whole-lung lavage.

Prospectus for the Future Sensitive and specific noninvasive methods are needed for the early identification of ILDs when attempts at preventing progression of lung fibrosis are likely to be more effective. IPF, the most common of the IIPs, is almost invariably fatal, and current treatment strategies are ineffective. Several clinical trials are examining the effectiveness of novel drugs in its treatment, including anti–TNF-α agents, endothelin receptor antagonists, and antioxidants. The National Institutes of Health have established the Idiopathic Pulmonary Fibrosis Clinical Research Network to accelerate discovery. However, much confusion about this disease remains in the community, and educational strategies will be needed to accelerate diagnosis. Sarcoidosis is another enigmatic disease that, when progressive, is largely unresponsive to current treatment strategies. Small studies suggest agents capable of immunomodulation might be useful, but further work is needed in this area. The advent of new technology able to evaluate genetic abnormalities related to

References Allen TC: Pulmonary Langerhans cell histiocytosis and other pulmonary histiocytic diseases: A review. Arch Pathol Lab Med 132:1171-1181, 2008. Collard HR, Schwarz MI: Diffuse alveolar hemorrhage. Clin Chest Med 25;583-592, 2004. Frankel SK, Cosgrove GP, Fischer A, et al: Update in the diagnosis and management of pulmonary vasculitis. Chest 129;452-465, 2006. Ianuzzi MC, Rybicki BA, Teirstein AS: Sarcoidosis. N Engl J Med 357:2153-2165, 2007. Joint statement of the American Thoracic Society (ATS) and the European Respiratory Society (ERS): American Thoracic Society/European Respiratory Society international multidisciplinary consensus classification of the idiopathic interstitial pneumonias. Am J Respir Crit Care Med 165:277-304, 2002.

disease has unveiled gene polymorphisms associated with IPF and sarcoidosis, among other ILDs. However, the true role of these abnormalities in causing the disease remains unclear, and the ability to exploit this information for early detection of disease is still limited. Interesting research is ongoing related to less common ILDs such as lymphangioleiomyomatosis, which affects women of childbearing age, and is focusing on the intracellular pathways that lead to cellular dysfunction in these disorders. Further work in this area and in the detection of the environmental hazards responsible for ILD is desperately needed. Until new and effective treatment strategies are generated, lung transplantation represents the only hope for an increasing number of patients with fibrosing ILDs. Therefore, efforts to extend life in lung transplant recipients are underway, particularly those targeting chronic rejection and bronchiolitis obliterans, the main cause of death in this population.

Leslie KO: Historical perspective: A pathologic approach to the classification of idiopathic interstitial pneumonias. Chest 128(5 Suppl 1):513S-519S, 2005. Limper AH: Drug-induced pulmonary disease. In Mason R, Broaddus VC, Murray JF, Nadel JA (eds): Murray and Nadel’s Textbook of Respiratory Medicine, 4th ed. Philadelphia, Elsevier Saunders, 2005, pp 1888-1908. Krymakaya VP: Smooth muscle-like cells in pulmonary lymphangioleiomyomatosis. Proc Am Thorac Soc 5:119-126, 2008. Noth I, Martinez FJ: Recent advances in idiopathic pulmonary fibrosis. Chest 132;637-650, 2007. Trapnell BC, Whitsett JA, Nakata K: Pulmonary alveolar proteinosis. N Engl J Med. 349:2527-2539, 2003. Wechsler ME: Pulmonary eosinophilic syndromes. Immunol Allergy Clin North Am 27:477-492, 2007.

Chapter

Pulmonary Vascular Disease Sharon Rounds

ulmonary vascular diseases are a heterogenous group of disorders with multiple causes. Pulmonary vascular disorders are caused by conditions that directly affect the pulmonary vessels, as in idiopathic pulmonary arterial hypertension (IPAH), or by disorders outside of the lung, as in pulmonary hypertension associated with lung disease and hypoxemia. The World Health Organization classification of pulmonary hypertensive disorders is presented in Table 19-1. The main complication of these disorders is the development of pulmonary hypertension, which is defined as mean pulmonary artery pressure over 25mmHg at rest or over 30 mm Hg with exercise. Factors that increase pul monary arterial pressure include cardiac output, left atrial pressure, blood viscosity, and (most importantly) loss of cross-sectional area of the vascular bed, which increases vascular resistance. The loss of a cross-sectional area may be the result of mechanical occlusion, loss of vessels, vascular remodeling, or vasoconstriction. Clinical manifestations of the disease may not be exhibited until late in the course of the disease. This delayed onset occurs because the pulmonary vasculature is a high-flow, low-resistance, highly compliant system with very high capacitance such that it can accept the entire output of the right ventricle with only slight increases in pressure—even when one half of the pulmonary vasculature is removed.

Pulmonary Thromboembolic Disease Pulmonary thromboembolic disease is a relatively common entity with an incidence ranging from 400,000 to 650,000 patients per year in the United States. Pulmonary thromboembolic disease is usually a complication of venous thrombosis. The deep veins of the femoral and popliteal systems of the lower extremities are most often affected, but right atrial, right ventricular, and upper extremity thromboses can also embolize to the lung. In view of this, predisposing factors for pulmonary embolism are the same as those

for venous thrombosis and include venous stasis, hypercoagulability, and endothelial injury. Congenital or acquired procoagulant disorders (e.g., activated protein C deficiency) are also considered predisposing factors. After a clot dislodges from the lower extremity circulation, it travels to the pulmonary circulation, where it can obstruct a branch of the pulmonary artery. The affected lung segment develops an increased ventilation-perfusion ). This increases overall dead space ventilation, ratio ( V Q which leads to an inefficient excretion of partial pressure of carbon dioxide in arterial blood (Paco2). In addition, blood flow is shifted from the obstructed site to other areas, thereby leading to which may include areas of low V Q, shunting and hypoxemia. Pulmonary infarction of the area distal to the occlusion is rare because of the redundancy of the pulmonary circulation and because of oxygenation of lung parenchyma by bronchial arteries and by alveolar oxygen.

CLINICAL PRESENTATION The classic presentation of acute pulmonary embolism includes acute shortness of breath accompanied by chest pain, hemoptysis, severe hypoxemia, and circulatory collapse as a result of shock. However, more often than not, the presentation is subtle, and the diagnosis might be difficult to make without a high level of suspicion, particularly in young individuals with otherwise healthy lungs. Dyspnea on exertion and atypical chest pain might be the only symptoms on initial presentation. Therefore, a careful history is paramount when evaluating patients for thromboembolic disease, especially those at high risk for this disorder as a result of stasis, malignancy, and previous history of venous thrombosis as well as other risk factors. The physical examination might reveal abnormalities in lung auscultation ranging from isolated crackles to diffuse wheezing. Pleural effusions might be underlying areas of dullness to percussion during the physical examination. Edema of the extremities, especially if the edema is asymmetrical, might point to venous thrombosis. In deep vein thrombosis, dorsiflexion of 241

242

Section IV—Pulmonary and Critical Care Medicine History and physical examination

Table 19-1 World Health Organization Classification of Pulmonary Hypertension Group I. Pulmonary Arterial Hypertension • Idiopathic (primary) • Familial • Related conditions, e.g., collagen vascular disease, portal hypertension, systemic-to-pulmonary shunts, HIV infection • Associated with significant venous or capillary involvement: pulmonary veno-occlusive disease and pulmonary-capillary hemangiomatosis • Persistent pulmonary hypertension of the newborn Group II. Pulmonary Venous Hypertension • Left-sided atrial or ventricular heart disease • Left-sided valvular heart disease Group III. Pulmonary Hypertension associated with Hypoxemia • Chronic obstructive pulmonary disease • Interstitial lung disease • Sleep-disordered breathing • Alveolar hypoventilation disorders • Chronic exposure to high altitude • Developmental abnormalities Group IV. Pulmonary Hypertension Due to Chronic Thrombotic Disease, Embolic Disease, or Both • Thromboembolic obstruction of proximal pulmonary arteries • Thromboembolic obstruction of distal pulmonary arteries • Pulmonary embolism (tumor, parasites, foreign material)

Suspected PE

Leg Doppler or venogram

DVT

· · V/Q scan

Spiral CT

Angiogram

Low Intermediate

High probability

PE confirmed

Figure 19-1 Tests commonly used in the evaluation of patients who may have pulmonary embolism (PE). Doppler ultrasound or venogram of the leg is useful to evaluate deep vein ) scans are most thrombosis (DVT). Ventilation-perfusion ( V Q useful when they are normal or show lesions highly suggestive of intravascular clot. Unfortunately, these findings are not the case in many patients, requiring further investigation. Spiral computed tomography (CT) has high sensitivity and specificity and allows for the evaluation of thoracic structures in addition to assessing the vasculature. Angiography is considered the gold standard, but it is often not needed if other noninvasive tests are used alone or in combination.

Group V. Miscellaneous • Sarcoidosis, pulmonary Langerhans cell histiocytosis, lymphangiomatosis, compression of pulmonary vessels (adenopathy, tumor, fibrosing mediastinitis)

the foot may cause calf pain as a result of stretching the calf muscles and deep veins (Homan sign). Signs of pulmonary hypertension and right ventricular strain, such as increased pulmonary component of the second heart sound or right ventricular heave, are not usually appreciated unless there is a massive pulmonary embolus or preexisting heart or lung disease.

EVALUATION In severe cases, arterial blood gas measurement may show acidemia, hypoxemia, and hypercapnia, but subtle changes such as mild alkalosis might be the only abnormalities. A normal Paco2 in a patient with tachypnea and presumably hyperventilation suggests increased dead space and, in the appropriate setting, might point to the diagnosis. However, a normal alveolar-arterial oxygen-tension gradient (A-aDo2) does not exclude acute pulmonary embolism. An elevated level of lactic dehydrogenase (LDH) might be the result of tissue infarction, but this test is also insensitive and nonspecific. Some have advocated the use of plasma D-dimer levels in patients who might have pulmonary thromboembolism, but these are not specific either because they are elevated in patients with several unrelated medical conditions such as congestive heart failure, chronic illness, and connective tissue disorders. The main usefulness of plasma D-dimer levels is its negative predictive value.

The electrocardiogram may show atrial tachyarrhythmias or evidence of right heart strain as evidenced by a new right bundle branch block, right ventricular strain pattern, and the SIQIITIII pattern that mimics inferior myocardial infarction. The chest radiograph is often normal but may show atelectasis, isolated infiltrates, or a small pleural effusion. Oligemia (Westermark sign), an abrupt cutoff of pulmonary vessels or enlarged central pulmonary arteries (Fleischer sign), and pleural-based area of increased opacity (Hampton hump) might also be noted. Independent of these findings, chest radiographs are not sensitive enough to diagnose pulmonary embolism. Three diagnostic methods are used for scan, chest the diagnosis of pulmonary embolism: the V Q computed tomography (CT), and pulmonary arteriography scan compares lung ventilation by (Fig. 19-1). The V Q radiolabeled tracer gas with lung perfusion by radiolabeled scan micro-occlusive particles. The usefulness of the V Q depends greatly on the pre-test probability of the disease, which, in turn, is dependent on the expertise of the clinician scan and his or her level of certainty. A high-probability V Q is characterized by lobar or multilobar perfusion defects that coincide with areas of normal or relatively normal ventilation and is more than 90% accurate in diagnosing pulmo scan shows no perfusion or nary embolism. A normal V Q ventilation defects and can exclude pulmonary embolism in essentially all cases. However, the test is less reliable when interpreted as low, intermediate, or indeterminate probability. Under such circumstances, pulmonary embolism is likely in between 4% and 66% of patients (Table 19-2), and further testing is necessary for an accurate diagnosis of pulmonary embolism.

Chapter 19—Pulmonary Vascular Disease Table 19-2 Pulmonary Embolism Likelihood Using Clinical Suggestion and Ventilation-Perfusion Scan Scan Result High Intermediate Low Normal All scans

Clinical Probability

80%-100% 96 66 40 0 68

20%-79% 0%-19% All 88 28 16 6 30

56 16 4 2 9

87 30 14 4 28

Data from The PIOPED Investigators: Value of ventilation/perfusion scan in acute pulmonary embolism: Results of the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED). JAMA;263:27532759, 1990.

Spiral CT angiography provides a noninvasive and sensitive way to evaluate for pulmonary emboli (Web Fig. 19-1). Pulmonary arteriography is the gold standard and should be considered in patients without contraindications to the procedure when other tests are inconclusive and a high likelihood of pulmonary embolism exists. Although complication rates related to this procedure are low, the complications are significant if developed, ranging from pulmonary hypertension and sudden death to idiosyncratic hypersensitivity reactions to dye. For this reason, many clinicians rely on a combination of interventions to arrive at the diagnosis, particularly when pulmonary tests are combined with tests that evaluate the deep veins of the lower extremities such as venography and Doppler ultrasound.

MANAGEMENT Pulmonary embolism is treated with supportive measures directed at sustaining organ function (e.g., fluid replacement for hypotension, mechanical ventilation for respiratory failure). To date, the only mechanical way to dislodge reliably a pulmonary artery clot is with surgical thromboembolectomy, a procedure with high mortality that requires a high level of expertise. Thromboembolectomy is only used for proximal clots that are long-standing (chronic thromboembolism syndrome). Consequently, medical treatments are preferred, and these are directed to prevent further clotting or to dissolve an existing clot. Anticoagulation with regular or low–molecular-weight heparin is recommended in patients without major contraindications to anticoagulation (e.g., upper gastrointestinal bleeding, hemorrhagic stroke). Their administration through subcutaneous injection appears to be as efficient as intravascular administration. The use of thrombolytic medications (e.g., tissue plasminogen activator) is usually reserved for patients with increased risk for mortality as a result of circulatory collapse caused by obstruction to the flow in large or multiple pulmonary vessels.

Idiopathic Pulmonary Arterial Hypertension IPAH is an uncommon disorder that is progressive and usually fatal without treatment. The median survival after

243

the diagnosis of the disease is about 3 years without treatment. Variables associated with poor survival include heart failure, Raynaud phenomenon, elevated right atrial pressure, significantly elevated mean pulmonary arterial pressure, and decreased cardiac index. The peak incidence of IPAH is between the ages of 20 and 45 years, and it affects women more frequently than men. The cause of IPAH is unknown. However, some cases occur in families, termed familial pulmonary arterial hypertension (FPAH). The genetic cause of FPAH has been determined and is due to mutations in bone morphogenetic protein receptor type 2 and related receptors in the transforming growth factor-β family. Some cases of pulmonary arterial hypertension are associated with other disorders, such as HIV infection, scleroderma, hepatic cirrhosis, and anorectic drug use (see Table 19-1). The histologic characteristics of IPAH are changes in both the arterial and venous systems. The arteries are more commonly affected, with changes in intima, media, and adventitia. There is medial vascular smooth muscle hypertrophy, adventitial thickening, and in situ thromboses of small pulmonary arteries. Plexogenic pulmonary arteriopathy is the classic pathologic finding in pulmonary arterial hypertension, consisting of medial hypertrophy, intimal proliferation and fibroelastosis, and necrotizing arteritis. The plexiform lesion is an abnormal proliferation of pulmonary endothelial cells with slitlike channels (Web Fig. 19-2). Like pulmonary thromboembolism, the clinical presentation of IPAH can be subtle. The usual symptoms are dyspnea on exertion or chest pains, not typical of angina pectoris. In more severe cases, patients may present with syncope on exertion caused by inability of the restricted pulmonary circulation to accommodate increased cardiac output with exercise. Chest radiographs may reveal prominent pulmonary arteries or right ventricular enlargement (Web Fig. 19-3). Pulmonary function tests are usually normal, with the exception of decreased diffusing capacity, reflecting the restricted circulation and decreased surface area available for gas exchange. Indeed, the diagnosis of IPAH is dependent on exclusion of other underlying heart or lung diseases that might result in pulmonary hypertension. Echocardiogram is useful to exclude heart diseases that increase pulmonary venous pressures (e.g., mitral valve stenosis). In addition, echocardiogram may reveal enlarged right atrial and right ventricular cavity size and encroachment of the interventricular septum on the left ventricle (Web Fig. 19-4). Furthermore, echocardiogram may be used to estimate the level of pulmonary artery systolic pressure. The definitive diagnosis of IPAH requires right heart catheterization with measurement of pulmonary artery pressures and resistance. Modern treatment of IPAH improves survival and includes drugs with vasodilator activity such as calcium channel blockers and prostacyclin. Because of the potential adverse effects of calcium channel blockers (decreased preload leading to acute hypotension), continuous intravenous prostacyclin is considered the most effective medical treatment. Other vasodilator drugs now available include endothelin receptor antagonists (e.g., bosentan) and drugs increasing cyclic guanosine monophosphate because of phosphodiesterase inhibition (e.g. sildenafil). These oral agents, plus inhaled or subcutaneous prostacyclin preparations, have dramatically enhanced treatment options for

244

Section IV—Pulmonary and Critical Care Medicine

patients with IPAH. In addition to effects on relaxing vascular smooth muscle constriction, vasodilator drugs also appear to stabilize or reverse vascular remodeling in IPAH. Other interventions include supplemental oxygen, anticoagulation, and judicious use of diuretic medications. Heartlung, double-lung, or single-lung transplantations have been performed in these patients with some success, but the overall 5-year survival rate in all patients undergoing lung transplantation is only 50%.

monary hypertension. For example, alveolar hypoxia causes intense pulmonary vasoconstriction. Long-standing hypoxia causes vascular remodeling that is similar to plexogenic pulmonary arteriopathy, but does not include in situ thromboses or formation of plexiform lesions (Web Fig. 19-5). Treatment of secondary pulmonary hypertension is directed at the underlying heart or lung disease. If hypoxia is present, home oxygen therapy should be used.

Secondary Pulmonary Hypertension

Cor Pulmonale

As shown in Table 19-1, pulmonary hypertension is also associated with other disorders that increase pulmonary venous pressure (e.g., mitral valve stenosis) and diseases of the lungs associated with hypoxemia (e.g., sleep apnea and chronic obstructive pulmonary disease. These conditions are frequently termed secondary pulmonary hypertension. Both vasoconstriction and vascular remodeling contribute to increased pulmonary vascular resistance in secondary pul-

It is now recognized that the most frequent cause of death in patients with IPAH is right ventricular failure, also termed cor pulmonale. Prolonged increased afterload causes the right ventricle to hypertrophy and then dilate. The interventricular septum shifts to the left, and filling of the left ventricle is decreased, with subsequent decreased cardiac output. Dilation of the right atrium causes atrial tachyarrhythmias and further decreased cardiac output. Treatment of cor pulmonale is directed at treatment of the underlying cause of pulmonary hypertension.

Prospectus for the Future Translational research has markedly enhanced understanding of the pathogenesis of pulmonary hypertensive disorders, and this has resulted in development of therapies that increase quality of life and improve mortality. There is increased appreciation of the role of increased vascular cell proliferation in the development of pulmonary vascular remodeling. In particular, abnormal proliferation of pulmonary endothelial cells and development of plexiform lesions have raised the suggestion that IPAH might be a disease of hyperproliferative pulmonary endothelium. In addition, little is understood regarding the

References Farber HW, Loscalzo J: Pulmonary arterial hypertension. N Engl J Med 351:16551665, 2004. Humbert M, Sitbon O, Simonneau G: Treatment of pulmonary arterial hypertension. N Engl J Med 351:1425-1436, 2004.

adaptive changes of the right ventricle to chronically increased afterload. Future investigations are needed to understand and better treat cor pulmonale. In contrast, less new information is available about pulmonary thromboembolic disease. Although new inhibitors of the coagulation cascade are currently under investigation, understanding of the mechanisms that lead to this illness has not dramatically changed during the past decade. Studies are needed in the area of genetic predisposition for thromboembolic disease as well as in vascular dysfunction leading to thrombus formation.

Newman JH, Phillips JA III, Loyd JE: Narrative review: The enigma of pulmonary arterial hypertension: New insights from genetic studies. Ann Intern Med 148:278-283, 2008. Tapson VF: Acute pulmonary embolism. N Engl J Med 358:1037-1052, 2008.

Chapter

Disorders of Respiratory Control Sharon Rounds and Matthew D. Jankowich

uring the transition between wakefulness and sleep, input from the behavioral control system decreases, the hypoxic drive to breathing is reduced, and the ventilatory response to partial pressure of carbon dioxide in arterial blood (Paco2) is diminished. These changes are most dramatic during rapid eye movement (REM) sleep. Sleep-disordered breathing refers to a diverse group of conditions in which these physiologic variations are heightened, resulting in abnormal respiratory function and fragmented sleep. Of the sleep-related disorders, sleep apnea has received the most attention. Apnea is defined as the complete cessation of airflow for 10 seconds or longer. Hypopnea is a significant decrease in airflow. Occasional episodes of apnea and hypopnea are expected during normal sleep, and their frequency increases with age. However, in patients with sleep apnea, the frequency and duration of the episodes are increased, leading to sleep fragmentation and to hypoxemia and hypercapnia. Upper airway obstruction (i.e., obstructive sleep apnea [OSA]) or decreased central respiratory drive (i.e., central sleep apnea) may be the cause of sleep apnea. In some patients, both disorders are present. Some studies suggest that the prevalence of sleep-disordered breathing may be as high as 9% in women and 24% in men, but prevalence levels depend on the definition used. Sleep-disordered breathing is usually defined as a respiratory disturbance index or frequency of abnormal respiratory events that number at least five episodes per hour of sleep. Higher prevalence estimates occur in the older adult population, with some studies showing more than 80% prevalence in older patients. Children are also affected, although less frequently (about 2%).

Obstructive Sleep Apnea OSA is the most common of the sleep apnea syndromes and is considered to affect close to 6% of middle-aged and older men; it is less common in women. In these patients, the upper airway relaxation that occurs during sleep is such that complete occlusion of the airway results, and, consequently,

cessation of airflow occurs. After variable periods of airway occlusion, the patient arouses, re-establishes muscle tone, and opens the airway. This vicious cycle is repeated many times during the night, resulting in recurring episodes of hypoxemia. During airway occlusion, sympathetic tone is increased, resulting in vasoconstriction and hypertension, which persists during the waking hours. Indeed, OSA is the most common identifiable cause of systemic hypertension. With airway occlusion, intrathoracic pressure becomes more negative with inspiration. Episodes of hypoxemia can be associated with bradycardia and cardiac arrhythmias. These events are believed to be linked mechanistically to the increased incidence of stroke and coronary artery disease in patients with OSA. An important physiologic consequence of airway occlusion is arousal from sleep, resulting in fragmented sleep. Because apneas are more frequent during REM sleep, patients complain of lack of refreshing sleep. Patients with OSA have an increased incidence of motor vehicle crashes, presumably related to somnolence while driving. Interestingly, patients with OSA display an increased incidence of diabetes mellitus and other manifestations of the metabolic syndrome. The cardiovascular complications of OSA appear to be at least partially reversible with treatment of OSA.

CLINICAL MANIFESTATIONS The diagnosis of OSA is suggested when patients complain of morning headaches, recurrent awakenings, and daytime somnolence that affects daytime activities, including driving. Complaints of snoring and gasping episodes may be elicited from sleeping partners. Difficulties in maintaining sleep as a result of frequent awakenings may lead to mood effects and decreased quality of life. Recent weight gain, sedatives and sleeping pills, or alcohol intake may heighten these symptoms. The primary risk factors for OSA are obesity (although variable) and abnormal upper airway anatomy caused by macroglossia, long soft palate and uvula, enlarged tonsils, or micrognathia. Increased neck diameter (>17cm in men and 245

246

Section IV—Pulmonary and Critical Care Medicine

>16cm in women) may also be noted. A narrow oropharynx as a result of a small pharyngeal opening or redundant soft tissue is often observed. Patients may be hypertensive and, in extreme cases, may show right-sided heart failure, which results as a consequence of prolonged episodes of hypoxemia and pulmonary vasoconstriction leading to pulmonary hypertension.

EVALUATION Chest radiographic images and pulmonary function testing are usually not helpful in the evaluation of patients with sleep apnea. In some cases, OSA is associated with the obesity-hypoventilation syndrome, which is characterized by significant obesity associated with chronic hypoventilation and hypoxemia (pickwickian syndrome). In such cases, arterial blood gases show hypoxemia and hypercapnia, and blood cell counts might suggest polycythemia. Although rare, hypothyroidism, acromegaly, and amyloidosis can cause or enhance OSA, and these conditions should be considered. A formal diagnosis requires overnight polysomnography during which continuous recordings of electrocardiographic and electroencephalographic tracings are made while the patient sleeps. In addition, airflow, oxygen saturation, and respiratory, eye, chin, and limb muscle movements are monitored and recorded. OSA is diagnosed in sleeping patients (confirmed by the electroencephalographic tracings) who develop cessation of airflow despite repeated muscular efforts to breathe (Web Fig. 20-1). These episodes may be accompanied by transient hypoxemia and cardiac arrhythmias. A score is derived from these data that defines clinically significant sleep apnea. Polysomnography will distinguish OSA from central sleep apnea, during which cessation of airflow is associated with halted respiratory movements. Polysomnography is also important to rule out other sleep disturbances, such as insomnia, narcolepsy, and parasomnias, as well as restless leg syndrome. Treatment of sleep apnea includes behavioral and medical approaches. When associated with obesity, weight loss should be enthusiastically encouraged. Avoidance of sedatives and alcohol is also important. Airway obstruction can be prevented with the use of continuous positive airway pressure (CPAP) provided through a tightly fitted mask. CPAP maintains positive airway pressure throughout expiration, thereby preventing collapse of the upper airway. The amount of pressure needed can be titrated, and oxygen can be added to further prevent hypoxemic episodes. CPAP is effective in most patients, but compliance with this technique is variable. Surgical removal of obstructing tonsils, adenoids, and polyps or uvulopalatopharyngoplasty may be useful in patients with specific anatomic abnormalities. A permanent tracheostomy may be necessary in severe cases when other approaches fail. However, in general, the surgical

approach to this disorder is limited to select patients only after CPAP has failed.

Other Disorders Related to Respiratory Control Central sleep apnea is a rare disorder. It predominates in men and is generally associated with normal body habitus. Patients may complain of daytime sleepiness and insomnia with frequent awakenings. This disorder is due to apnea or hypopnea, resulting from decreased central respiratory drive, and may be a consequence of central nervous system injury (i.e., central apnea may be the result of a structural abnormality of the brainstem) or idiopathic. Affected individuals may hypoventilate even while awake, although they are capable of normal voluntary breaths. During sleep, frequent apnea is common. In patients with obstructive lung disease, increased work of breathing eventually makes it difficult to maintain sufficient ventilation to maintain normal levels of Paco2. When ventilatory capacity declines, hypoventilation causes Paco2 to increase; the kidneys respond by retaining bicarbonate to keep arterial blood pH at normal levels. These patients appear to have normal ventilatory drive, but they lack the ability to increase minute ventilation to meet increased metabolic demand. This characteristic is observed in certain patients with chronic bronchitis who exhibit the classic description of the “blue bloater.” Lower brainstem and upper pontine lesions may cause central hyperventilation. However, this disorder rarely occurs in the absence of other physiologic or chemical abnormalities. Hepatic cirrhosis and extreme anxiety are all causes of central hyperventilation. Pregnancy can also cause hyperventilation and is thought to be caused by elevated levels of progesterone and other hormones that increase central actions. Apneustic breathing consists of sustained inspiratory pauses, resulting from damage to the mid-pons, most commonly caused by basilar artery infarction. Biot respiration or ataxic breathing is a haphazardly random pattern of sleep and is characterized by shallow breaths; a disruption of the respiratory rhythm generator in the medulla causes this sign. The regular cycling of crescendo-decrescendo tidal volumes, separated by apneic or hypopneic pauses, characterizes Cheyne-Stokes respiration. Patients with this disorder usually have generalized central nervous system disease or congestive heart failure. Heart failure prolongs circulatory times, causing a delay between changes in blood gases at the tissue level and the arrival of those changes at the brainstem chemoreceptors. This delay sets up a cycle of gradual increase to hyperventilation, followed by gradually decreasing ventilation to apnea, and then a repetition of the cycle. Recent studies suggest that OSA and Cheyne-Stokes respiration not only are consequences of congestive heart failure but also contribute to progression of CHF.

Chapter 20—Disorders of Respiratory Control

247

Prospectus for the Future With more than 5% of the population in the United States suffering from sleep-disordered breathing, and with the recognition that these disorders may contribute to systemic illnesses such as hypertension and cardiovascular disorders, interest in early diagnosis and treatment of disorders of respiratory control is growing. In view of the high incidence and potential health consequences of sleep-disordered breathing, physicians must be on the lookout for this condition. The increased inci-

References Arzt M, Bradley TD: Treatment of sleep apnea in heart failure. Am J Resp Crit Care Med 173:1300, 2006. Caples SM, Gami AS, Somers VK: Obstructive sleep apnea. Ann Intern Med 142:187-197, 2005. Somers VK, White DP, Amin R, et al: Sleep apnea and cardiovascular disease. J Am Coll Cardiol 52:686, 2008.

dence of OSA parallels that of obesity in the United States, a public health problem that has been associated with asthma and increased risk for death. It is highly likely that genetic predisposition to OSA accounts for increased incidence in some families. Sleep medicine is an emerging clinical and research area that will continue to receive great attention in the coming decade.

Stephen GA, Eichling PS, Quan SF: Treatment of sleep disordered breathing and obstructive sleep apnea. Minerva Med 95:323-336, 2004. White DP: Pathogenesis of obstructive and central sleep apnea. Am J Resp Crit Care Med. 172:1363, 2005. Young T, Skatrud J, Peppard PE: Risk factors for obstructive sleep apnea in adults. JAMA. 201:2013, 2004.

Chapter

Disorders of the Pleura, Mediastinum, and Chest Wall F. Dennis McCool

Pleural Disease The pleura is a thin membrane that covers the entire surface of the lung as well as the inner surface of the rib cage, diaphragm, and mediastinum. There are two pleural membranes: the visceral pleura, which covers the lung; and the parietal pleura, which lines the rib cage, diaphragm, and mediastinum. A layer of mesothelial cells lines both pleural surfaces. The closed space in between the surface of the lung and the chest cavity is referred to as the pleural space. A small amount of fluid normally resides in this space and forms a thin layer between the pleural surfaces. Pleural fluid serves as a lubricant for the visceral and parietal pleura as they move against each other during inspiration and expiration. The blood vessels in the visceral pleura are supplied from the pulmonary circulation and have greater hydrostatic pressure than the blood vessels in the parietal pleura, which are supplied by the systemic circulation. The pressure within the pleural space is subatmospheric during quiet breathing. Fluid is filtered from the higher-pressure vascular structures into the pleural space. The normal fluid turnover is about 10 to 20 mL per day with 0.2 to 1 mL remaining in the pleural space. Pleural fluid usually contains a small amount of protein and a small number of cells that are mostly mononuclear cells. Although both the parietal and visceral pleura contribute to pleural fluid formation, most of the fluid results from filtration of the higher-pressure vessels supplying the parietal pleura. After the fluid enters the pleural space, it is drained from the pleural space by a network of pleural lymphatics located beneath the mesothelial mono layer. The lymphatics originate in stomas on the parietal pleural surface. Under abnormal circumstances, fluid can accumulate in the pleural space. Factors that promote the entry of fluid into the pleural space include an increase in systemic venous pressure, an increase in pulmonary venous pressure, an increase in permeability of pleural vessels, or a 248

reduction in pleural pressure. Conditions that increase hydrostatic pressure can be seen with congestive heart failure; changes in pleural membrane permeability can be seen in varied inflammatory states; and a reduction in pleural pressure can be seen with atelectasis. Occasionally, micro vascular oncotic pressure may be sufficiently reduced to promote fluid entry into the pleural space in patients with hypoalbuminemia. Factors that block lymphatic drainage and interfere with the egress of fluid from the pleural space include central lymphatic obstruction or obstruction of lymphatic channels at the pleural surface by tumor.

PLEURAL EFFUSION Pleural effusion is the accumulation of fluid in the pleural space. Pleural effusions are generally detected by chest radiography; however, the volume of fluid in the pleural space needs to exceed 250 mL to be visualized on a chest radiograph. When an effusion is present, there is blunting of the costophrenic angle on a posteroanterior chest film; this is a fluid meniscus that can be detected posteriorly also on the lateral chest radiograph, and occasionally fluid can be demonstrated in either the minor or major fissures (Web Figs. 21-1 and 21-2). Changes in the contour of the diaphragm may signify a subpulmonic effusion. A decubitus chest radiograph can be obtained to determine whether the fluid is free-flowing or loculated. A computed tomography (CT) scan of the chest provides better definition of the pleural space than plain radiography. Chest CT is particularly useful in differentiating pulmonary parenchymal abnormalities from pleural abnormalities, defining loculated effusions, distinguishing between atelectasis and effusion, and distinguishing loculated effusion from lung abscess (Web Fig. 21-3). The edge of a lung abscess usually touches the chest wall and forms an acute angle whereas that of an empyema is usually an obtuse angle.

Chapter 21—Disorders of the Pleura, Mediastinum, and Chest Wall

249

Table 21-1 Pleural Effusions Transudates Congestive heart failure Hypoalbuminemia Nephrotic syndrome Malnutrition Cirrhosis Intra-abdominal fluid Ascites Peritoneal dialysis

Exudates Infection Empyema Parapneumonic Malignancy Primary lung cancer Lymphoma Metastatic cancer Pulmonary embolism and infarction Collagen vascular disease

Systemic lupus erythematosus Rheumatoid arthritis Intra-abdominal pathologic abnormalities Pancreatitis Subphrenic abscess Complications of abdominal surgery Meigs syndrome Urinothorax

Trauma Hemothorax Chylothorax Ruptured esophagus Miscellaneous Myxedema Uremia Asbestosis Lymphedema Drug-induced lupus Dressler syndrome

From Light RW, Macgregor MI, Luchsinger PC, etal: Pleural effusions: The diagnostic separation of transudates and exudates. Ann Intern Med 77:507513, 1972.

Thoracentesis is a procedure in which fluid is aspirated from the pleural space. Ultrasound or a CT scan can be used to help direct the thoracentesis catheter into collections of fluid that are otherwise difficult to drain. Analysis of pleural fluid may provide a definitive diagnosis; however, even without a definitive diagnosis, pleural fluid analysis can be useful in excluding other possible causes of disease such as infection. Classifying pleural effusions as transudates or exudates greatly assists with the differential diagnosis. The general approach to pleural effusions is outlined in Web Figure 21-4.

TRANSUDATES Effusions that accumulate due to changes in osmotic and hydrostatic forces usually have low protein states and are considered transudates (Table 21-1). Congestive heart failure is the most common cause of a transudate. With heart failure, the effusions are typically bilateral. If the effusion is unilateral, it involves the right hemithorax in most instances. Effusions due to heart failure are related to dysfunction of the left side of the heart, not the right side of the heart. Transudative effusions also may be seen in cirrhosis, nephrotic syndrome, myxedema, pulmonary embolism, superior vena cava obstruction, and peritoneal dialysis. In patients with cirrhosis, the effusions are often right sided, and the mechanism may be related to flow from the peritoneal space across diaphragmatic defects into the pleural space. Transudative effusions are typically small and rarely require drainage to improve symptoms.

EXUDATES Exudative effusions occur when there is an alteration in vascular permeability and can be observed in inflammatory states, with infection, or with neoplasm. To distinguish an exudate from a transudate, one of three criteria must be fulfilled: (1) pleural fluid–to–serum protein ratio is greater than 0.5; (2) pleural fluid–to–serum lactate dehydrogenase (LDH) ratio is greater than 0.6; and (3) pleural fluid LDH is greater than two thirds the upper limit of normal (Table 21-2). When all three criteria are met, the sensitivity, specificity, and predicted value exceed 98% for an exudative effusion. Measuring pleural fluid cholesterol may also help

Table 21-2 Differentiation of Exudative and Transudative Pleural Effusions Exudate

Transudate

Protein Pleural and serum protein LDH

>3g/dL >0.5

125 beats per minute); tachypnea (>30 breaths per minute); high fever (>38.3° to 40° C); hypotension (systolic blood pressure < 90mmHg); hypoxia (Sao2 < 90% or Pao2 < 60 mm Hg); multilobar involvement on chest radiograph; and identification of highrisk pathogens such as gram-negative organisms and S. aureus. For hospitalized patients, initial therapy for community-acquired pneumonia usually includes a cephalosporin such as ceftriaxone or cefuroxime, with or without a macrolide. Antibiotic treatment should be given as soon as possible because mortality can increase even after a short delay (>8 hours) in receiving appropriate antibiotics. Sputum and blood cultures should be obtained before instituting antibiotic therapy.

Adapted from Modai J: Empiric therapy of severe infections in adults. Am J Med 88:12S-17S, 1990.

Nosocomial Pneumonia community-acquired pneumonia is designed to cover this organism (see later). M. pneumoniae is a slow-growing, facultative anaerobic organism that accounts for 25% to 60% of all atypical pneumonias. M. pneumoniae is a common cause of pneumonia in patients between the ages of 5 and 35 years who may initially exhibit upper respiratory tract symptoms, pharyngitis, and bullous myringitis. Dry cough, fever, gastrointestinal symptoms, headache, and myalgias are common. Uncommon complications include cold agglutinin–induced hemolysis, hepatitis, erythema multiforme, the syndrome of inappropriate antidiuretic hormone, pericarditis, myocarditis, and neurologic abnormalities. The chest radiograph may show fine interstitial reticulonodular infiltrates in patients, which are often relatively asymptomatic. The diagnosis is based on clinical and epidemiologic features. Acute and convalescent serologic findings are required to confirm the diagnosis, but are not helpful during the acute illness. Other common causes of community-acquired pneumonia are C. pneumoniae and Haemophilus influenzae. Patients with co-morbid conditions and those older than 65 years are also at risk for pneumonia from Legionella species, Staphylococcus aureus, and gram-negative organisms. Anaerobic infection should be considered when large amounts of oropharyngeal secretions are aspirated and in patients with chronic infections in the gingivodental crevice. Diagnostic tests for community-acquired pneumonia should include a chest radiograph and complete blood count. The role of routine sputum and blood cultures in this setting is controversial. The recommended treatment for community-acquired pneumonia is a course (7 to 10 days) of a macrolide antibiotic (erythromycin, clarithromycin, or azithromycin). If there are co-morbidities such as chronic heart or lung disease, an extended spectrum fluoroquinolone (e.g., levofloxacin, moxifloxacin, or gemifloxacin) or a β lactam (amoxicillin) plus a macrolide should be used. The choice of treatment should also be influenced by local antibiotic resistance patterns.

Nosocomial pneumonia is pneumonia that occurs after hospitalization. Nosocomial pneumonia is subdivided into three categories: hospital-acquired pneumonia (HAP); ventilator-associated pneumonia (VAP); and health care– associated pneumonia (HCAP). HAP is defined as pneumonia that occurs 48 hours or more after admission and that did not appear to be incubating at the time of admission. VAP is a type of HAP that develops more than 48 to 72 hours after endotracheal intubation. HCAP is defined as pneumonia that occurs in a nonhospitalized patient with extensive health care contact. This includes recent hospitalization, residence in a nursing home or other long-term care facility, and recent intravenous therapy. These patients should be considered at high risk for resistant organisms and therefore inappropriate for routine, empirical therapy for community-acquired pneumonia. Nosocomial pneumonia is the second most common infection in hospitalized patients and the most common infection in the intensive care unit. The pathogenesis of nosocomial pneumonia is based on colonization of the oropharynx and stomach with virulent pathogens and the subsequent aspiration of these organisms into the lower respiratory tract. Gastric colonization by gram-negative organisms is enhanced by neutralization of gastric acidity. In the first 5 days of hospitalization, H. influenzae, S. pneumoniae, and S. aureus are often isolated. After this time, pneumonia is often caused by Pseudomonas aeruginosa, S. aureus, anaerobic microbes, Acinetobacter species, and various gram-negative enteric bacilli. This finding has important therapeutic implications because these organisms are more commonly associated with multidrug antibiotic resistance. Treatment is dependent on combined chemotherapy with β lactam antipseudomonal penicillin or cephalosporin, together with an aminoglycoside or a quinolone. Vancomycin is added if methicillin-resistant S. aureus is suspected. A more definitive identification of organisms and their sensitivity to antibiotics is often sought in these patients using more invasive measures, including endotracheal

256

Section IV—Pulmonary and Critical Care Medicine

aspirate in intubated patients or flexible fiberoptic bronchoscopy. However, the best predictor of patient outcome with nosocomial pneumonia appears to be adequacy of the initial empirical antibiotic regimen.

Complications of Pneumonia Parapneumonic effusion is a neutrophilic exudative effusion adjacent to a lung with pneumonia (Web Fig. 22-2). It can resolve with antibiotics alone or can require drainage in addition to antibiotics. As a pneumonia progresses, inflammatory pulmonary liquid leaks into the pleural space, first appearing as an uncomplicated effusion. At this point, the effusion will resolve with antibiotic therapy alone. As bacteria and inflammatory cells follow, the inflammatory process is marked by anaerobic metabolism, cytokine production, fibrin deposition in the pleural space, and thickening of the pleura. There is no universally accepted definition of empyema. However, most clinicians include in the term empyema all pleural effusions that are grossly purulent or contain microorganisms identified by a positive Gram stain or culture. Empyema must always be treated with pleural drainage, usually by a chest thoracostomy tube. Highly inflammatory parapneumonic effusions may behave as if they are infected, although microorganisms are never identified. There have been effusions described as “complicated” parapneumonic effusions, and these are identified clinically by a pH of less than 7.1 and a glucose level of less than 40 mg/dL. It is important to recognize that complicated effusions generally require drainage in addition to antibiotic therapy. The major risk factor for the development of lung abscess is aspiration resulting in a more indolent, polymicrobial infection, usually involving both aerobes and anaerobes. Conditions predisposing patients to aspiration, such as alcoholism, seizures, or stroke, are associated with an increased incidence of lung abscess. The presence of poor dentition increases the anaerobic bacterial load in the mouth and thus the likelihood of infection after an aspiration event. For aspiration-related infection, the antibiotic chosen should reflect the predominance of anaerobes. In trials of empirical therapy for lung abscess, clindamycin showed superiority over penicillin, probably because the incidence of penicillinresistant anaerobes in lung abscesses is 15% to 20%. Antibiotics should be continued for 6 weeks, and drainage should be reserved for very large abscesses or failure to resolve with antibiotics.

existed. An estimated 1.87 million individuals die from TB each year, and the global case-fatality rate was 23%, with 50% in some African countries with high HIV rates. In the United States, TB increased at an alarming rate in the early 1990s as a result of the surge of HIV infection, drug abuse, inner-city poverty, and homelessness. TB infection occurs when aerosolized, contaminated droplets (expectorated by a diseased person) are inhaled by another individual and the droplet or droplet nuclei reaches an alveolus. This is almost always a latent infection, called latent tuberculosis infection (LTBI). If the innate immune system of the host fails to eliminate the latent infection, the bacilli proliferate inside alveolar macrophages and kill the cells. The infected macrophages produce cytokines and chemokines that attract other phagocytic cells, including monocytes, other alveolar macrophages, and neutrophils, which eventually form a nodular granulomatous structure called the tubercle. If the bacterial replication is not con trolled, the tubercle enlarges, and the bacilli enter the local draining lymph nodes. This leads to lymphadenopathy, a characteristic manifestation of primary TB. The lesion produced by the expansion of the tubercle into the lung parenchyma and lymph node involvement is called the Ghon complex. The bacilli continue to proliferate until an effective cell-mediated immune (CMI) response develops, usually 2 to 6 weeks after infection. Failure by the host to mount an effective CMI response and tissue repair leads to progressive destruction of the lung. Bacterial products, tumor necrosis factor-α, macrophage antimicrobial effector molecules such as reactive oxygen intermediates (ROI) and reactive nitrogen intermediates (RNI), and the contents of cytotoxic cells (granzymes, perforin) all can contribute to the development of caseating necrosis that characterizes a tuberculous granuloma (Fig. 22-1). If mycobacterial growth is unchecked, the bacilli may spread hematogenously to produce disseminated TB. Miliary

Mycobacterium Tuberculosis Infection Infection with Mycobacterium tuberculosis, an aerobic, nonmotile, acid-fast rod with niacin production, causes TB. In 1997, the World Health Organization Global Surveillance and Monitoring Project estimated 8 million new cases per year of TB, including 3.5 million cases of infectious pulmonary disease. In addition, 16.2 million cases of the disease

Figure 22-1 Necrotizing granuloma in lung infected with Mycobacterium tuberculosis.

Chapter 22—Infectious Disease of the Lung TB is a disseminated form with lesions resembling millet seeds. Bacilli can also spread mechanically by erosion of the caseating lesions into the lung airways. It is at this point that the host becomes infectious to others. If untreated, 80% of patients will die. Others will develop chronic disease or recover spontaneously. The chronic disease is characterized by repeated episodes of spontaneous healing with fibrotic changes around the lesions and tissue breakdown. Healing by complete spontaneous eradication of the bacilli is rare. Reactivation TB results when the persistent bacteria in a host suddenly proliferate. Only 5% to 10% of patients with no underlying medical problems who become infected develop reactivation disease in their lifetime. Although immunosuppression is clearly associated with reactivation TB, it is not clear what host factors specifically maintain the infection in a latent state for many years and what triggers the latent infection to become overt. The diagnosis of latent TB infection is dependent on a positive tuberculin test, which does not necessarily indicate active disease, but only previous infection. The standard Mantoux test is an intradermal injection of 0.1mL (5 tuberculin units) of purified protein derivative (PPD) tuberculin in the skin of the forearm. The injection site is evaluated 48 to 72 hours later. The reading is based on the diameter of the indurated or swollen area. Patients are at high risk for developing active TB early after tuberculin conversion, and thus treatment is recommended for LTBI. The risk for active disease is 5% within 2 years of exposure and another 5% per year thereafter. HIV patients are an exception and have a 40% risk for active disease within several months of conversion. The current recommendations as to what constitutes a positive PPD test take into account the degree of clinical suspicion of LTBI (Table 22-2). Treatment of patients suspected of having active disease includes at least four drugs—isoniazid, 5 mg/kg per day; rifampin, 10 mg/kg per day; ethambutol, 5 to 25 mg/kg per day; pyrazinamide, 15 to 30mg/kg per day—and should Table 22-2 Prophylaxis against Tuberculosis in Adults

PPD 5mm

10mm

15mm

Prophylaxis Indicated Regardless of Age Close contacts recently diagnosed with TB HIV positive or HIV risk factors Fibrotic changes on chest radiograph Diabetes mellitus Immunosuppression Hematologic malignancy Injection drug use Renal failure Malnutrition PPD increased >15mm within 2yr

Prophylaxis Indicated if 10mm within 2yr Native of highprevalence country High-risk ethnic minorities Residents and staff of long-term care facilities No risk factors

HIV, human immunodeficiency virus; PPD, purified protein derivative of tuberculin; TB, tuberculosis.

257

be considered before a formal diagnosis is made. Factors suggesting active disease include exposure to active TB, pulmonary symptoms, and cavitary disease on imaging studies. If the diagnosis of TB is confirmed, the drugs are continued for 2 months, barring adverse reactions to drug therapy. After 2 months, the regimen can be tailored, depending on drug-sensitivity studies, and continued for another 4 months with at least two active drugs. Rates of drug-resistant TB are increased in certain populations (e.g., recent immigrants from high TB areas, homeless people). Resistance is detected in 9% of patients who have not received previous therapy and in 22.8% of those with prior treatment. In patients with drug-resistant TB, treatment should include at least three drugs that have not been administered before and to which the organism is susceptible in vitro. Treatment should continue for at least 18 to 24 months. Direct observation of therapy is recommended to ensure compliance.

Pneumocystis Pneumonia Pneumocystis jiroveci, formerly called pneumocystis carinii (PCP), is an opportunistic fungus that occurred mainly in malnourished premature infants and in adults with hematologic malignancy undergoing chemotherapy in the pre– acquired immunodeficiency syndrome (AIDS) era. However, its incidence rose significantly in the late 1980s and 1990s in patients with AIDS with low CD4+ lymphocyte counts (30 breaths per minute), mental deterioration (e.g., impaired judgment, confusion, hallucinations, somnolence), or hemodynamic instability (e.g., bradydysrhythmias or tachydysrhythmias, hypotension) usually require intubation and mechanical ventilation. In the latter circumstances, waiting for arterial blood gas determinations is not necessary and could dangerously delay therapy. Although arterial blood gas evaluation is crucial when determining the need for mechanical ventilation in the patient with respiratory failure, the patient’s clinical status will ultimately dictate the course of action.

Mechanical Ventilation Mechanical ventilation primarily uses the principals of positive pressure ventilation. Air is forced into the central airways, increasing central airway pressure. Air follows the pressure gradient from the central airways to the alveoli, which inflates the lungs. As the lungs inflate and the device stops forcing air into the central airways, the intra-alveolar pressure increases and central airway pressure decreases. Exhalation occurs when the air follows the newly reversed pressure gradient from the alveoli to the central airways. The principal benefits of mechanical ventilation during respiratory failure are improved gas exchange and decreased work of breathing. Mechanical ventilation improves gas 259

260

Section IV—Pulmonary and Critical Care Medicine

Q) exchange by improving ventilation-perfusion ratio ( V matching. The improved V Q matching is primarily a consequence of decreased physiologic shunting. Altered lung mechanics (e.g., increased airways resistance, decreased compliance) and increased respiratory demand (e.g., metabolic acidosis) increase the work of breathing. The ventilatory muscles and diaphragm can tire while trying to maintain the elevated work of breathing, resulting in respiratory failure. Mechanical ventilation can assume some or all of the increased work of breathing, allowing the ventilatory muscles to recover from their fatigue. Deteriorating gas exchange, unresponsive to conservative measures, and respiratory distress are the most common reasons for mechanical ventilation in patients with acute respiratory failure.

NONINVASIVE MECHANICAL VENTILATION Although intubation and mechanical ventilation are usually the preferred options in respiratory failure that is considered reversible, noninvasive positive-pressure ventilation (NPPV) has proved useful in selected patients. NPPV refers to positive-pressure ventilation delivered through a noninvasive interface (nasal mask, facemask, or nasal plugs), rather than an invasive interface (endotracheal tube, tracheostomy). Its use has become more common as benefits are increasingly recognized. Selecting patients for NPPV requires careful consideration of its indications and contraindications. Generally speaking, a trial of NPPV is worthwhile in patients with acute cardiogenic pulmonary edema or hypercapnic respiratory failure due to COPD who do not require emergent intubation and do not have contraindications to NPPV. Contraindications to NPPV include cardiac or respiratory arrest; inability to cooperate, protect the airway, or clear secretions; severely impaired consciousness; facial surgery, trauma, or deformity; anticipated prolonged duration of mechanical ventilation; and recent esophageal anastomosis.

INVASIVE MECHANICAL VENTILATION Once a decision to intubate is made, an experienced operator should expeditiously perform intubation. Complications of intubation are usually related to prolonged hypoxemia as a result of delays in the procedure, but they also include vomiting and aspiration of gastric contents, trauma to the vocal cords, bleeding, pneumothorax, cardiac arrhythmias, and cardiac arrest. Once inserted, the endotracheal tube should be secured and its position assessed by examining for breath sounds, followed by chest radiography for confirmation. Direct visualization through a bronchoscope is occasionally needed for successful intubation. A ventilator should be available before the procedure is begun so that mechanical ventilation can start as soon as the endotracheal tube is secured. Initial ventilator settings may vary, but typically they include a ventilator mode, fraction of inspired oxygen (Fio2) of 1 (or 100%), respiratory rate set at 10 to 12 breaths per minute, and tidal volume of 400 to 600mL. The adequacy of the ventilator settings needs to be determined with repeated arterial blood gas levels and the clinical evaluation of the patient. Persistent cyanosis, pallor, diaphoresis, and restlessness may suggest that the tube is misplaced or that

the ventilator settings are insufficient to ventilate the patient appropriately. Positive end-expiratory pressure (PEEP) might be required in patients with refractory hypoxemia. PEEP prevents the premature collapse of the alveoli during matching, leading to improved Q expiration and improves V oxygenation. Once the settings are adjusted to maintain relatively normal levels of arterial blood gases (pH, 7.3 to 7.45; Pao2 > 60mmHg; Pco2, 30 to 45mmHg), attention should be given to developing a maintenance plan that will secure adequate oxygenation and ventilation until the cause of the respiratory failure is treated and the failure is reversed. This plan should include assessment of the need for sedation, appropriate strategy of mechanical ventilation, supportive measures to achieve hemodynamic stability, nutritional assessment, and therapies targeting the initial injurious process that triggered the respiratory failure (e.g., pneumonia, pulmonary embolism, asthma, shock). Most patients require sedation to diminish discomfort and to decrease the work of breathing, but it should be administered carefully because sedation is often accompanied by a decrease in blood pressure. Commonly used modes of ventilation are determined by the duration of inspiration, which can be limited by volume, pressure, flow, or time. During volume-limited ventilation, inspiration ends after delivery of a preset tidal volume. Airway pressure is variable during volume-limited ventilation and is related to respiratory system compliance, airway resistance, and tubing resistance. Assist control (AC), continuous mandatory ventilation (CMV), and synchronized intermittent mandatory ventilation (SIMV) are examples of modes of ventilation that can be volume limited. CMV has a set rate and set tidal volume which does not allow spontaneous breathing by the patient. Patient-ventilator asynchrony is a big problem, and therefore CMV is rarely used. The assist-control mode of ventilation is similar to CMV in that there is a set rate and tidal volume, but this mode allows the patient to initiate additional spontaneous breaths. When the machine senses that the patient is attempting to take a breath, it delivers the selected tidal volume. SIMV is similar to assist-control in that a set rate and tidal volume are selected. The patient is also able to generate a spontaneous breath. However, this spontaneous breath may have a very small tidal volume, yet entail significant work of breathing. Consequently, this mode of mechanical ventilation is seldom used except when weaning patients from mechanical ventilation. The pressure control mode of ventilation uses machine breaths that are pressure cycled, not volume cycled. With pressure control ventilation, the pressure to be used for each breath is ordered. If the patient attempts a spontaneous breath, a machine breath at the designated pressure is delivered. This may be helpful in limiting airway pressures in patients with bronchospasm or stiff lungs because it limits the risk for pneumothorax (barotrauma). Because tidal volumes may vary, the pressure control mode must be titrated carefully at the bedside to determine the proper pressure settings. Pressure support ventilation is used only for spontaneously breathing patients. The inspiratory and expiratory pressures are selected, and there are no mandatory machine breaths. Patients find this to be a more comfortable mode of mechanical ventilation. However, pressure support ventilation

Chapter 23—Essentials in Critical Care Medicine should only be used for patients with a stable respiratory drive (not sedated heavily) and stable lung compliance. Pressure support ventilation is typically used for patients who are weaning from mechanical ventilator support. Pressure-regulated volume control, airway pressure release ventilation, and high-frequency ventilation are newer modalities that are increasingly used in clinical practice.

SETTINGS Numerous settings need to be considered when mechanical ventilation is initiated. These include tidal volume, respiratory rate, trigger mode and sensitivity, fraction of inspired oxygen, PEEP, flow rate, and flow pattern. The appropriate initial tidal volume depends on numerous factors, most notably the disease for which the patient requires mechanical ventilation. The tidal volume can then be increased or decreased incrementally to achieve the desired pH and arterial carbon dioxide tension (Paco2). Generally speaking, large tidal volumes can cause barotrauma or volutrauma, which increases the risk for ventilator-associated lung injury. Therefore, tidal volume should not be increased without considering effects on airway pressure or the likelihood of ventilator-induced lung injury. An optimal method for setting the respiratory rate has not been established. Once the tidal volume has been established, the respiratory rate can be incrementally increased or decreased to achieve the desired pH and Paco2, while monitoring auto-PEEP. Patients who are breathing spontaneously will set their own respiratory rates in all modes of ventilation except CMV. The lowest possible fraction of inspired oxygen (FiO2) necessary to meet oxygenation goals should be used. This will decrease the likelihood that adverse consequences of supplemental oxygen will develop, such as absorption atelectasis, accentuation of hypercapnia, airway injury, and parenchymal lung injury. PEEP is generally added to prevent end-expiratory alveo matching and Q lar collapse. This generally improves V arterial oxygenation and allows reduction in Fio2, thereby reducing the risk for oxygen toxicity. However, elevated levels of applied PEEP can have adverse consequences, such as reduced preload (decreases cardiac output), elevated plateau airway pressure (increases risk for barotrauma), and impaired cerebral venous outflow (increases intracranial pressure). The optimal PEEP is that which enhances oxygenation without lung hyperinflation and decreased blood pressure. Respiratory therapists typically also adjust the inspiratory flow rate, flow pattern, and amount of negative pressure required to “trigger” a mechanical ventilator breath. If these ventilator settings are not adjusted with due consideration of the patient’s respiratory mechanics, two common problems can occur: asynchrony and auto-PEEP. Patientventilator asynchrony occurs if the phases of breaths delivered by the ventilator do not match the breathing pattern of the patient. Patient-ventilator asynchrony can cause dyspnea, increase the work of breathing, and prolong the duration of mechanical ventilation. It is detected by careful observation of the patient and examination of the ventilator waveforms. Generally, the abnormality that is most readily apparent is

261

failure of the ventilator to trigger a breath when the patient makes an inspiratory effort. Auto-PEEP is usually seen when patients are not fully emptying their lungs during expiration before the initiation of the next breath. This is known as stacking breaths or generating auto-PEEP. This is particularly worrisome in patients who have exacerbations of COPD or status asthmaticus requiring mechanical ventilation. In ventilated patients, auto-PEEP may cause barotrauma or hemodynamic collapse owing to high intrathoracic pressures preventing blood return to the right ventricle.

Weaning from Mechanical Ventilation The complications of endotracheal intubation and mechanical ventilation are many; trauma to the lung from high ventilator pressures (barotrauma), volutrauma and ventilator-induced lung injury, and pneumonia are the most significant. Therefore, the patient who is mechanically ventilated must be treated aggressively and monitored carefully. Weaning from mechanical ventilation should be considered when the original insult that caused respiratory failure has cleared, especially if the patient is awake and cooperative and shows no signs of respiratory or hemodynamic instability. Weaning is usually not attempted if requirements for oxygen supplementation remain high (Fio2 > 0.5). Conventional parameters that determine whether weaning is possible include negative inspiratory force, vital capacity, tidal volume, respiratory rate, and minute ventilation (Table 23-1). Unfortunately, the strength of these parameters lies in the ability to predict failure to wean rather than in the ability to predict successful spontaneous breathing. A better way to assess weaning capability is to engage the patient in a short weaning trial during which support from the ventilator is diminished. This trial can be achieved by allowing the patient to breathe oxygen for 1 hour without providing supporting pressure. Another strategy is to decrease the pressure generated by the ventilator during a trial of continuous positive airway pressure (CPAP). The patient is monitored for any signs of distress or hemodynamic instability, and arterial blood gas levels are measured to determine the effectiveness of spontaneous ventilation. If the patient tolerates the trial, extubation may be indicated, depending on the patient’s clinical status and his or her underlying condition. In general, Table 23-1 Conventional Weaning Parameters Parameters NIF (cm of water) VC (mL/kg) VT (mL/kg) RR (breaths/min) VE (L/min) Rapid shallow breathing index (RSBI) (RR/VT)

Weanable Values

Normal Ranges

10 10 3cm diameter or tumor of any size with any of the following characteristics: Invasion of visceral pleura Atelectasis of less than entire lung Proximal extent at least 2cm from carina T3—Tumor of any size with any of the following characteristics: Invasion of chest wall Involvement of diaphragm, mediastinal pleura, or pericardium Atelectasis involving entire lung Proximal extent within 2cm of carina T4—Tumor of any size with any of the following: Invasion of mediastinum Invasion of heart or great vessels Invasion of trachea or esophagus Invasion of vertebral body or carina Presence of malignant pleural or pericardial effusion Satellite tumor nodule(s) within same lobe as primary tumor Nodal Involvement (N) N0—No regional node involvement N1—Metastasis to ipsilateral hilar and/or ipsilateral peribronchial nodes N2—Metastasis to ipsilateral mediastinal and/or subcarinal nodes N3—Metastasis to contralateral mediastinal or hilar nodes or ipsilateral or contralateral scalene or supraclavicular nodes Metastasis (M) M0—Distant metastasis absent M1—Distant metastasis present (includes metastatic tumor nodules in a different lobe from the primary tumor) Stage Groupings of TNM Subsets Stage Stage Stage Stage

IA IB IIA IIB

T1 T2 T1 T2 T3

N0 N0 N1 N1 N0

M0 M0 M0 M0 M0

Stage IIIA Stage IIIB Stage IV

T3 N1 M0 T1-3 N2 M0 Any T N3 M0 T4 Any N M0 Any T Any N M1

Adapted from Greene FL, Page DL, Fleming ID: AJCC Cancer Staging Manual, 6th ed. New York, Springer, 2002.

disease is limited. Combinations of cisplatin and etoposide are the standard chemotherapeutic regimen. Although chemotherapy and radiation often produce a dramatic response and sometimes are curative for limited disease, relapse is typical, and subsequent treatments are less effective. Prophylactic cranial radiation may be performed. Surgery is the only curative therapy for NSCLC and is indicated for patients with stage I or II NSCLC who are operative candidates. Lobectomy (or greater) extent of resection is considered superior to more limited resection such as a wedge resection. Adjuvant chemotherapy is appropriate for patients with stage II disease. For patients with stage IIIA lung cancer, the optimal treatment strategy remains unclear, in part because of the heterogeneity of patients in this group. In general, these patients are not candidates for surgery alone, if at all, and a treatment plan should be developed for each patient in a multidisciplinary setting. In stage IIIB,

269

surgery may rarely be indicated for some T4 N0-1 M0 tumors. However, most patients with stage IIIB NSCLC are not surgical candidates, and 5-year survival is poor for this group. Combined, ideally concurrent, chemotherapy and radiotherapy are preferable to radiotherapy alone in patients with stage IIIB NSCLC. In stage IV, chemotherapy is recommended because it improves survival and provides palliation for symptoms. Targeted molecular therapies are an area of active interest in lung cancer treatment. Highlighting the challenges of these new targeted therapies, bevacizumab, a humanized monoclonal antibody against vascular endothelial growth factor, was associated with hemoptysis, which was sometimes fatal, in patients with squamous cell carcinoma in an early-stage trial. However, bevacizumab improved survival when added to a standard platinum-based chemotherapeutic regimen in patients with nonsquamous NSCLC. Erlotinib, a tyrosine kinase inhibitor targeting the activity of the epidermal growth factor receptor, is approved for the second-line treatment of metastatic NSCLC. Targeting the epidermal growth factor receptor appears to have benefit in particular patient groups, such as women, never-smokers, and Asians, harboring particular receptor mutations.

Special Circumstances SOLITARY PULMONARY NODULE A solitary pulmonary nodule (SPN) is a single, rounded lesion in the lung that is 3cm or less in diameter. Although these lesions are commonly lung cancers in certain patient populations, the differential diagnosis of this radiographic finding is broad and includes many malignant and benign etiologies. In addition to primary lung cancer (with adenocarcinoma the most common type to present as an SPN (see Web Fig. 24-12), other malignant etiologies include bronchial carcinoid tumors and metastatic foci from extrapulmonary malignancies (most common sources include malignant melanoma, sarcoma, colon, kidney, breast, and testicle). Benign etiologies include benign tumors of the lung (hamartomas (Web Fig. 24-14), infectious granulomas (from fungal diseases including histoplasmosis and coccidioidomycosis as well as mycobacterial disease), lung abscess, vascular abnormalities (arteriovenous malformation), rounded atelectasis, and pseudotumor (pleural fluid trapped within a fissure). When confronted with a patient with a solitary pulmonary nodule, determining the likelihood of malignancy is critically important because early resection of malignant nodules is usually curative, whereas resection of benign nodules exposes the patient to an unnecessary risk for surgery. Diagnostic evaluation includes consideration of certain clinical features, including patient age and risk factors. The probability of an SPN being malignant increases with patient age. However, even in younger individuals (7 METs) Carry 24lb up eight steps Carry objects that weigh 80lb Outdoor work (shovel snow, spade soil) Recreation (ski, basketball, squash, handball, jog or walk 5mph)

• Emergent major operations, especially in elderly patients • Aortic and other major vascular (endovascular and nonendovascular) surgery • Noncarotid peripheral vascular surgery • Prolonged surgery associated with large fluid shift and/or blood loss

275

Intermediate Risk

Moderate (activities requiring >4 but 70, 3.5g/1.73m2 per day), edema, and hypoalbuminuria Oval fat bodies, coarse granular casts

Poststreptococcal glomerulonephritis

Nephrotic syndrome With bland urine sediment (pure nephrotic) Asymptomatic urinary abnormalities Tubulointerstitial nephropathy Acute renal failure Rapidly progressive renal failure Tubular defects

Isolated proteinuria (36 to 48 hours), the brain generates compounds that raise the intracellular osmolality and thereby minimize cell shrinkage. This process is metabolically driven and is slow to return to normal. Thus, rapid correction of plasma osmolality may lead to a shift of water to the relatively hypertonic intracellular compartment and may result in brain edema. As a general rule, hypernatremia should be corrected over 48 hours at a rate not exceeding 0.5mEq/L per hour, or 12mEq/L per day.

Disturbances in Potassium Balance The human body contains about 3500 mEq of potassium. With a normal concentration of 3.5 to 5 mEq/L, the ECF contains about 70mEq of potassium, or only 2% of totalbody stores. In response to a dietary potassium load, rapid removal of potassium from the extracellular space is necessary to prevent life-threatening hyperkalemia. For example, in the absence of a homeostatic mechanism, if a person ingests 50 mEq of dietary potassium in a single meal (the average daily American diet contains 100 to 120 mEq of potassium per day), the serum potassium might rise to 7mEq/L (assuming an extracellular volume of 14L with a baseline serum potassium of 4 mEq/L). Thus, the initial adaptation to a potassium load is the rapid redistribution of potassium from the extracellular space to the intracellular space. Various hormones, including insulin, aldosterone, and catecholamines, cause movement of potassium into cells. The acid-base status of the patient is another determinant of the serum potassium concentration because potassium moves across cell membranes driven by pH gradients between the cell and the ECF compartments. The greatest effect on the serum potassium concentration is associated with metabolic acidosis involving mineral acids. The cellular permeability to the anions of the mineral acids is low; consequently, the basolateral membrane is hyperpolarized, provoking potassium movement into the blood. By contrast, metabolic acidosis caused by organic acids, such as lactic acid and keto acids, does not cause hyperkalemia. The anions of these acids are relatively permeable and accompany hydrogen into the cell. This situation diminishes the electrochemical gradient favoring potassium efflux.

Although these mechanisms affect the distribution of potassium between the fluid compartments, other mechanisms are necessary to maintain overall potassium balance. People ingest about 100mEq of potassium daily, the bulk of which is eliminated by the kidneys. Increased potassium excretion results from enhanced distal nephron potassium secretion by the principal cells of the connecting tubule and collecting duct into the tubular lumen down an electrochemical gradient. Factors that enhance this gradient promote potassium secretion. These factors include the rate of distal tubular flow, the distal delivery of sodium, the presence of poorly reabsorbable anions in the tubular fluid, and stimulation by aldosterone. The ratio of extracellular to intracellular potassium establishes the resting membrane potential of the cell. Hence hyperkalemia or hypokalemia is associated with alteration of the resting membrane potential, which accounts for most of the symptoms and findings in these disorders.

DIAGNOSTIC APPROACH A careful history with emphasis on the patient’s diet and use of medications and laxatives should be obtained. Spurious hyperkalemia and hypokalemia must be excluded. In addition to serum electrolytes and magnesium, urine electrolytes and urine osmolality should be obtained. The next step should be to determine whether abnormal renal potassium handling is involved in the genesis of the disorder. This state may be determined by measuring the 24-hour urine potassium excretion. In extrarenal hyperkalemia, renal potassium excretion should be more than 200mEq/day, and if hypokalemia is caused by extrarenal losses, the renal potassium excretion should be less than 20mEq/day.

HYPERKALEMIA The ratio of intracellular to extracellular potassium concentration is the major determinant of the resting potential of the cell membrane. As the extracellular potassium concentration increases, the cell membrane is partially depolarized, the sodium permeability is diminished, and the ability to generate action potentials is decreased. In muscle tissue, this change accounts for muscle weakness and paralysis. In the heart, hyperkalemia exhibits as changes in the electrocardiogram. These changes include peaked T waves, decreased amplitude or the absence of P waves, wide QRS complexes, sinus bradycardia, and conduction defects. A pathophysiologic approach to the causes of hyperkalemia is outlined in Figure 28-6. Vigorous phlebotomy techniques can result in lysis of red blood cells, a process that releases intracellular potassium into the serum sample. Thrombocytosis (>1 × 106/µL) and leukocytosis (>60,000/µL) may also be associated with spurious hyperkalemia. These disorders can be diagnosed rapidly by determining the plasma and serum concentrations of potassium. True hyperkalemia is present if these values differ by less than or equal to 0.2mEq/L. Chronic renal insufficiency does not cause hyperkalemia unless it is advanced, with a GFR ranging from less than 10 to 15mL per minute. Thus, hyperkalemia in chronic renal insufficiency is usually caused by a distal nephron defect in potassium secretion rather than by the impaired GFR, as

Chapter 28—Fluid and Electrolyte Disorders

313

Hyperkalemia

Pseudohyperkalemia • Hemolysis • Thrombocytosis • Leukocytosis

Redistribution • Acidosis • ↓Insulin • β-Adrenergic blockade • Arginine infusion • Succinylcholine • Digitalis overdose (massive) • Periodic paralysis

Impaired renal K+ excretion* (TTKG 60,000 cells/µL), resulting from active uptake of potassium by white blood cells from the serum. True hypokalemia is caused by redistribution, extrarenal potassium loss, poor intake, or renal potassium losses. Because only 2% of total-body potassium is distributed in the extracellular compartment, serum potassium measurements may not accurately reflect the total-body stores. In fact, hypokalemia can occur in the presence of normal total-body potassium stores. This state occurs when potassium shifts from the extracellular space to the intracellular space. Excess circulating catecholamines, insulin administration, and alkalosis are the major causes of redistribution of potassium from the extracellular space to the intracellular space. Redistribution hypokalemia is particularly important in the clinical setting of myocardial infarction and exacerbation of chronic obstructive pulmonary disease. These patients are especially prone to arrhythmias because excess catecholamines (in response to stress or

314

Section VI—Renal Disease Hypokalemia

Normal acid-base

Redistribution • Catecholamine excess • Alkalosis • Hypokalemic periodic paralysis • Insulin administration • Barium poisoning

Extrarenal losses • Decreased intake (e.g., anorexia nervosa) • Laxative abuse

Metabolic acidosis

Extrarenal losses (urine K+ 2) • Renal tubular acidosis • Organic acidosis (lactic and ketoacidosis) • Carbonic anhydrase inhibitors

Urine K+ 20 mEq/L

High blood pressure • Hyperaldosteronism • Essential hypertension with diuretic use • Hypercortisolism • Apparent mineralocorticoid excess (licorice ingestion, Liddle syndrome)

Figure 28-7 Diagnostic approach to hypokalemia. TTKG, transtubular potassium gradient.

inhaled β2 agonists) cause potassium shifts in the setting of total-body potassium depletion from frequent diuretic usage. In patients with hypokalemia, the acid-base status, the presence or absence of hypertension, and measurement of urinary chloride and potassium are helpful in narrowing the diagnostic possibilities. In patients with diuretic abuse (usually patients with eating disorders), the urine sodium and chloride concentrations are high in the presence of metabolic alkalosis, a profile similar to that of Bartter syndrome, which is a rare genetic defect usually seen in adolescents with other neurologic abnormalities and caused by reduced activity of the sodium, potassium, and chloride (NKCC2) co-transporters in the thick ascending limb (see later). In this setting, a urine screen for diuretics may be necessary to make the diagnosis. In comparison, patients with surreptitious vomiting have a low urinary chloride concentration. Patients who abuse laxatives have low urine sodium and chloride concentrations, with metabolic acidosis or normal acid-base status. Glycyrrhizic acid, the active ingredient in licorice, blocks 11β-dehydrogenase, an enzyme that inactivates glucocorticoids, and its inhibition results in unregulated activation of the mineralocorticoid receptors in the distal nephron. Determination of serum magnesium should always be performed in a patient with hypokalemia. Hypokalemia that is associated with hypomagnesemia is resistant to therapy unless concomitant magnesium deficiency is corrected. Given the factors that determine transmembrane potassium shifts, the net potassium deficit may be difficult to

calculate. An estimate for a 70-kg man based on the serum concentration is a 100- to 200-mEq deficit in total-body potassium when the serum concentration decreases from 4 to 3mEq/L. At less than 3mEq/L, every 1-mEq/L decrease in the serum concentration of potassium reflects an additional 200- to 400-mEq deficit in total-body potassium. Hypokalemia should be treated with oral potassium supplementation. Intravenous potassium administration should only be used in urgent situations, such as in patients with arrhythmias or digitalis toxicity, and intolerance to oral formulations in patients with adynamic ileus. The rate of intravenous potassium administration generally should not exceed 10 mEq per hour; only under electrocardiographic monitoring, the potassium administration rate can be increased up to 20 mEq per hour. Hypokalemia associated with long-term diuretic therapy may be treated with the addition of a potassium-sparing diuretic.

Bartter and Gitelman Syndromes A major advance in the study of inherited disorders of salt-wasting syndromes has been the demonstration that Bartter and Gitelman syndromes result from mutation of specific ion transport proteins expressed by cells of the distal nephron. The dysfunction of the thiazide-sensitive sodiumchloride co-transporter (NCCT) in Gitelman syndrome, and the bumetanide-sensitive sodium-potassium-chloride

Chapter 28—Fluid and Electrolyte Disorders

315

Table 28-6 Various Bartter Syndromes Gene name Protein name Major symptoms Seizures Urine Ca2+ Urine Mg2+ Nephrocalcinosis

Type 1

Type 2

Type 3

Type 4

Type 5

SLC12A1 NKCC2 Polyuria, hypocalcemia Dehydration ↑ ↓ +++

KCNJ1 ROMK As for type 1 — ↑ ↓ +++

CLCNKB CLCNKB Variable Mild to severe ↑ ↓ ±

BSND Barttin As for type 1 + Deafness ↑ ↓ −

CASR CaR — — ↑ ↓ +++

Ca2+, calcium; Mg2+, magnesium.

Table 28-7 Known Pathophysiology of Gitelman Syndrome Loss of functional mutation: NCCT NaCl wasting Secondary hyperaldosteronism: K wasting Cellular hyperpolarization secondary decreased Cl− entry ↑ Apical ECaC entry ↑ Basolateral Na/Ca exchange Net effect: hypocalciuria Mg wasting: uncertain mechanism

(NKCC2) co-transporter in Bartter syndrome cause salt wasting, extracellular volume depletion, secondary hyperaldosteronism, and hypokalemia. The various characteristics of the five different Bartter syndromes and the Gitelman syndromes are shown in Tables 28-6 and 28-7. Stated briefly, Bartter syndrome is a genetically heterogeneous disease. Based on molecular genetic studies, five different subtypes of Bartter syndromes can be distinguished. Type I, or neonatal Bartter syndrome, is caused by loss of function mutations in the sodium-potassium-chloride co-transporter NKCC2. NKCC2 is encoded by the SCL12A1 gene on chromosome 15. This sodiumpotassium-chloride co-transporter is expressed in the apical cell membranes of the thick ascending limb of the loop of Henle (TAL), and normally accounts for about 30% of total reabsorption of sodium filtered by the glomerulus. Patients with this syndrome present early in life with a severe systemic disorder characterized by marked sodium and potassium wasting, polyhydramnios, and significant hypercalciuria and nephrocalcinosis. Prostaglandin synthesis and excretion are significantly increased and may account for much of the systemic symptoms. Bartter syndrome type II is due to loss of function mutation of the KCNJ1 gene on chromosome 11, encoding the inward rectifier voltage-dependent potassium channel ROMK. The potassium channel ROMK is localized in the apical membrane of TAL but is also expressed in the cortical collecting duct. In the TAL, the potassium flow through this channel into the renal tubule is necessary for NKCC2 activity, which needs adequate luminal potassium supply. In the cortical collecting duct, this channel is also involved in the excretion of dietary potassium. In this syndrome, an aberrant ROMK channel leads to malfunction of the NKCC2 co-transporter and results in salt wasting, high tubular flow, and distal potassium wasting. Bartter syndrome type III is

caused by loss of function mutations of the CLCNKB gene on chromosome 1 encoding a chloride channel protein CLC-Kb. This protein is expressed in the basolateral cell membranes of the TAL and is responsible for the reabsorption of sodium chloride in the TAL. The renal salt wasting in Bartter type III syndrome is less severe than types I and II. Recently, Bartter syndrome type IV was found to be caused by loss of function mutation of the BSND gene (Bartter syndrome and sensorineural deafness) on chromosome 1p31. The BSND gene encodes barttin, a protein expressed in the basolateral membrane of the TAL. Barttin is the β subunit of the ClC-Kb chloride channel; it is necessary for ClCKb delivery to the plasma membrane; and in the cochlea, it co-localizes with chloride channels ClCKa and ClCKb. These patients present also with sodium, potassium wasting, and impaired cochlear function and deafness. Recently, the presence of hypokalemic alkalosis due to Bartter syndrome type V was found in patients with autosomal dominant hypocalcemia. In this disease, hypocalcemia was related to gain of function mutation of the calcium-sensing receptor (CaSR). The CaSR is heavily expressed at the basolateral membrane of the TAL, where it is thought to play an important inhibitory role in regulating the transcellular transport of sodium, chloride, and calcium. Activation of the basolateral CaSR in the TAL reduces apical K+ channel activity, which induces a Bartter-like syndrome. Genetic activation of the CaSR by these mutations is also expected to increase urinary calcium excretion by inhibiting the generation of the lumen positive potential difference that drives paracellular calcium transport in the TAL. To date, Gitelman syndrome appears to be molecularly homogeneous. Although a loss of function mutation of the SLC12A3 gene in chromosome 16q13 has been identified as one of the causes, recent genetic characterization of Gordon syndrome, a disease with clinical features opposite to Gitelman syndrome, suggests the possibility that similar mirror-image mutations and possibly more than one, might account for Gitelman syndrome. The SLC12A3 gene encodes the renal thiazide-sensitive sodium-chloride co-transporter NCCT. NCCT is responsible for the sodium reabsorption in the distal tubule, which accounts for about 7% of the total filtered sodium. Although Gitelman syndrome is a milder disorder than Bartter syndrome, patients do report significant morbidity related to muscular symptoms, fatigue, and increased risk for cardiac arrhythmias in patients having a prolonged QT interval. Although plasma renin activity is increased, renal prostaglandin excretion is not elevated, another feature that distinguishes Gitelman syndrome from Bartter syndrome.

316

Section VI—Renal Disease

A major phenotypic difference between Bartter and Gitelman syndromes involves urinary calcium excretion. The hypercalciuria of Bartter syndrome is believed to result largely from dysfunction of thick ascending limb cells. Calcium absorption, which is passive and paracellular, is driven by lumen-positive transepithelial voltage that is generated by NKCC2 co-transport and luminal K+ recycling. When NKCC2 co-transport is reduced or blocked by loop diuretics or genetic abnormality, the lumen positivity declines, and calcium reabsorption declines. What is not entirely clear is why the loss of luminal positivity does not lead to increased magnesium excretion. In addition to this mechanism, increased distal delivery of NaCl, as a result of dysfunction of the TAL raises intracellular chloride concentration, which in turns inhibits apical calcium channel of distal convoluted tubule (DCT) cells, further contributing to calcium retention and nephrolithiasis. In contrast to patients with Bartter syndrome, patients with Gitelman syndrome invariably demonstrate hypocalciuria. The hypocalciuria of Gitelman syndrome resembles the clinical beneficial effects of DCT diuretics (thiazides and others) to reduce urinary calcium excretion. The mechanisms of hypocalciuria in Gitelman syndrome are well established. First, mild contraction of the ECF volume increases calcium reabsorption along the proximal tubule. Second, reduction in NaCl entry to DCT cells stimulates transepithelial calcium transport. When apical Na and Cl entry into DCT cells is inhibited, because of either diuretic treatment or genetic disease, the intracellular chloride concentration declines. Lower intracellular Cl activity hyperpolarizes the cell and activates calcium entry through the distinctive apical calcium channels ECaC and CaT2, expressed in the DCT cells. Because calcium movement from lumen to cell must be balanced, the increased calcium entry to DCT cells stimulates calcium efflux through the basolateral Na+/Ca+ exchanger and the Ca-ATPase. Therefore, the resultant effect is the development of hypocalciuria. Although the pathogenesis of calcium disorders in Bartter and Gitelman syndromes is relatively clear, only recently has the pathogenesis of magnesium disorders associated with these syndromes been clarified. Gitelman syndrome is associated with severe hypomagnesemia, whereas Bartter syndrome is not. Recent observations in genetic disorders of hypomagnesemia, as well as clinical observations in patients undergoing chemotherapy with anti–epidermal growth factor (EGF) antibodies have helped to clarify the molecular mechanisms involved in magnesium transport. Magnesium is absorbed along the entire nephron, but the predominant site for reabsorption is along the distal tubule. In the medullary and cortical TAL, magnesium, along with calcium, is reabsorbed through a charge-selective paracellular path. Mutations in claudin-16 (also called paracellin) and claudin19 cause severe hypomagnesemia and nephrocalcinosis. In the DCT, a specific channel, the transient receptor potential melastatin, subfamily 6 channel, TRPM6, mediates magnesium reabsorption. Mutations in the EGF gene, which is expressed in the distal tubule, cause hypomagnesemia, and anti-EGF antibodies induce hypomagnesemia. The latter observations suggest a regulatory role of EGF in the reabsorption of magnesium. Knockout of the thiazide-sensitive sodium chloride co-transporter and inhibition of this transporter with thiazides cause hypomagnesemia and reduce

TRPM6 expression. Thus, reduced expression of the TRPM6 channel is the most likely explanation of the hypomagnesemia seen in Gitelman syndrome.

Disturbances in Acid-Base Balance Most metabolic processes occurring in the body result in the production of acid. The largest source of endogenous acid production is from the complete catabolism and oxidation of glucose and fatty acids ultimately to carbon dioxide and water. Pulmonary ventilation excretes the volatile acid produced by such cellular respiration, about 22,000 mEq of hydrogen daily, as carbon dioxide. Cellular metabolism of sulfur-containing amino acids, the oxidation of phosphoproteins and phospholipids, nucleoprotein degradation, and the incomplete combustion of carbohydrates and fatty acids result in the formation of nonvolatile acids. These processes produce about 1 mEq/kg body weight of hydrogen daily. Nonvolatile acid excretion is effected through the kidney. The primary factors regulating alteration in the rate of minute ventilation are changes in cerebrospinal fluid and arterial blood pH. The normal concentration of hydrogen in arterial blood is 40 mEq/L, equal to a pH of 7.40. This concentration is maintained relatively constant despite variations in the endogenous and exogenous acid inputs. Circulating and intracellular buffers acutely neutralize an acid load. The capacity of these buffering systems is limited, however, and would be quickly depleted by normal endogenous acid production. Mechanisms for excreting acid must therefore be effective to regenerate these buffers to maintain acid-base homeostasis.

RENAL HYDROGEN ION EXCRETION The kidney contributes to acid-base homeostasis by the reclamation of 4500mEq of bicarbonate filtered at the glomerulus daily and by the generation of new bicarbonate. This renal bicarbonate generation is given by the equation for net acid excretion (NAE): E NAE = ENH4+ + E TA − EHCO3− where ENH4+ + ETA is the rate of ammonium and titratable acid excretion, respectively, and EHCO3−1 is the rate of bicarbonate excretion. Although not difficult, measurement of NAE is not routinely done, and thus analysis of acid-base disorders has relied on indirect but more readily available measurements. The principal process of renal bicarbonate generation is accomplished by urinary acidification and ammonia generation. Ammonia is generated from glutamine and secreted into the tubule fluid by the proximal tubule epithelium. Acidification is accomplished by proton secretion by distinct transporters uniquely distributed along the nephron. These transporters were highlighted in Chapter 26, and this chapter discusses how they modify the tubular fluid pH. The acidification profile of tubular fluid is shown in Figure 28-8. Tubule fluid pH falls slightly along the proximal tubule by

Chapter 28—Fluid and Electrolyte Disorders

Table 28-8 Systemic Approach to the Analysis of Acid-Base Disorders

% Proximal tubule

0.2 0 0.2

% Distal Ureteral convolution urine + 0 20 40 60 80 100 0 50 100 ∆pH ∆pH − −

0 0.2

0.4

0.6

0.8

1.0

1.0 1.2 9 Nondiuretic rats

317

1.4 1.6 1.8 2.0

Figure 28-8 pH changes along the rat nephron.

the operation of the sodium-hydrogen exchanger type 3 (NHE3) located at the luminal membrane. This proton secretion is linked to bicarbonate reabsorption, and nearly 90% of the filtered load of bicarbonate is reabsorbed from the tubule fluid during its course along the proximal tubule. The proximal tubule is a high-capacity bicarbonate reabsorption system whose rate results from the presence of carbonic anhydrase in the luminal membrane. Carbonic anhydrase in the luminal membrane rapidly catalyzes the dehydration of carbonic acid to carbon dioxide and water and thus restricts the fall in tubule fluid pH along the proximal tubule and prevents the development of too steep a pH gradient into which sodium-hydrogen exchange takes place. The distal tubule, on the other hand, lacks a luminal carbonic anhydrase and has a limited capacity to reabsorb bicarbonate. However, given that the bicarbonate concentration reaching these sites is low and the principal buffers in the tubule fluid along these sites are ammonia and phosphate ions, the continued operation of the distal proton pumps lowers the prevailing urinary pH to sometimes 1000fold below the pH of the initial filtrate. Along these sites, however, the full operation of the kidney in acid-base homeostasis is observed as the urinary buffers are titrated by secreted protons from pumps distributed along the distal nephron and excreted into the urine. The rate of bicarbonate generation is not fixed and responds to changes in volume and electrolyte status, hormones, and acid-base parameters. Proximal bicarbonate reabsorption is increased during volume depletion, by elevation in the partial pressure of carbon dioxide (Pco2), as seen in chronic respiratory acidosis, and by hypokalemia. Conversely, volume expansion or the reduction of the Pco2 lowers the proximal tubular reabsorptive rate for bicarbonate. Aldosterone and ambient Pco2 affect the rate of distal nephron hydrogen ion secretion.

ASSESSMENT OF ACID-BASE STATUS A systematic approach to assessing acid-base status consists of several steps, as summarized in Table 28-8. The initial step

Assess the accuracy of the acid-base parameters using the Henderson equation (H+ = 24 × [ PaCO2 HCO3− ) or the Henderson-Hasselbalch equation (pH = 6.1 + log[HCO3− 0.03] × Paco2). Obtain a good history and perform a complete physical examination, looking for clues to a particular acid-base disturbance. Calculate the serum anion gap: Na+− ( HCO3− + Cl−). Identify the primary acid-base disturbance and assess whether a simple or mixed acid-base disturbance is present. Examine serum electrolytes and ancillary laboratory data. Measure urine pH and urine electrolytes, urine urea nitrogen, and glucose to calculate the urine anion gap (Na+ + K+− Cl−) or urine osmolal gap (measured osmolality − [2(Na+ + K+) + [urea nitrogen/2.8] + [glucose/18]).* *Measurement of urine Na+ and Cl− and urine pH should be obtained when metabolic alkalosis is present. Measurement of urine electrolytes, urine glucose, and urine pH should be obtained when an element of normal anion gap metabolic acidosis is present.

is to obtain arterial and venous blood samples to measure blood pH and Pco2, as well as serum electrolytes to determine the nature of the acid-base disturbance. Validation of the internal consistency of the calculated and measured bicarbonate should be carried out. Based on the pH, Pco2, and serum bicarbonate, a minimum diagnosis should be established. Next, a measurement of the compensatory response and the anion gap should be performed. If the compensation of a primary acid-base defect is inappropriate, then a mixed acid-base disorder is considered (Table 28-9). The anion gap is useful in the diagnostic approach to metabolic acidosis. When an organic acid, such as lactic acid, is added to the ECF compartment, the bicarbonate concentration falls as the acid is buffered. The anion gap increases as the organic base is accumulated. Quantitatively, the increase in anion gap should be equivalent to the decrease in bicarbonate concentration. Thus, by adding the difference between the calculated and normal anion gap to the prevailing bicarbonate concentration, an estimate of the starting bicarbonate concentration can be made. An abnormally elevated initial bicarbonate concentration indicates concomitant metabolic alkalosis. After establishing the nature of the acid-base disorder and whether it is complex or simple, examination of the urine pH and urine anion or osmolal gap can also provide useful information.

METABOLIC ACIDOSIS Metabolic acidosis is characterized by a decrease in the serum bicarbonate concentration. This decrease occurs either by excretion of bicarbonate-containing fluids or by utilization of bicarbonate as a buffer of acids. In the latter instance, the nature of the base may affect the electrolyte composition. Thus, considering metabolic acidosis by means of the anion gap is convenient (Table 28-10). Metabolic acidoses with a normal anion gap is most commonly caused by extrarenal losses of bicarbonate, as occurs in diarrheal diseases, but may also be caused by abnormally high renal excretion of bicarbonate and by the addition of

318

Section VI—Renal Disease

Table 28-9 Nature of Adaptive Response to Primary Acid-Base Disorders Primary Acid-Base Disturbance

Initiating Mechanism

Metabolic acidosis

↓ plasma HCO3−

↓ Paco2 of 1.0-1.3mmHg for each 1mEq/L fall in plasma HCO3− Paco2 = (1.5 × HCO3− ) + 8 + 2

Metabolic alkalosis

↑ Plasma HCO3−

Respiratory acidosis

↑ Paco2

Respiratory alkalosis

↓ Paco2

↑ Paco2 of 0.4 to 0.7mmHg for each 1mEq/L rise in plasma HCO3− Acute: ↑ plasma HCO3− of 1mEq/L for every 10-mm Hg rise in Paco2 Chronic: ↑ plasma HCO3− of 3.5mEq/L for every 100-mmHg fall in Paco2 Acute: ↓ plasma HCO3− of 2mEq/L for every 10-mmHg fall in Paco2 Chronic: ↓ plasma HCO3− of 4-5mEq/L for every 10-mm Hg fall in Paco2

Secondary Physiologic Response

Note: Rare cases with Paco2 values greater than 55mmHg have been reported.

Table 28-10 Causes of Metabolic Acidosis Normal Anion Gap Bicarbonate losses Extrarenal Small bowel drainage Diarrhea Renal Proximal renal tubular acidosis Carbonic anhydrase inhibitors Primary hyperparathyroidism Failure of bicarbonate regeneration Distal renal tubular acidosis Aldosterone deficiency Addison disease Hyporeninemic hypoaldosteronism Aldosterone insensitivity Interstitial renal disease Aldosterone antagonists Ureteroileostomy (ileal bladder) Acidifying salts Ammonium chloride Lysine or arginine hydrochloride Diabetes mellitus (recovery phase) Wide Anion Gap Reduced excretion of acids Renal failure Overproduction of acids Ketoacidosis Diabetic Alcoholic Starvation Lactic acidosis Toxin ingestion Methanol Ethylene glycol Salicylates Modified from Andreoli TE: Disorders of fluid volume, electrolyte, and acid-base balance. In Wyngaarden JB, Smith LH Jr, Bennett JC (eds): Cecil Textbook of Medicine, 19th ed. Philadelphia: WB Saunders, 1992, p 523.

substances yielding hydrochloric acid, as when arginine hydrochloride is administered. The urinary anion gap is defined as follows: Urinary anion gap = (sodium + potassium ) − chloride The equation provides an approximate index of urinary ammonium excretion, as measured by a negative urinary anion gap. Thus, a normal renal response would be a negative urinary anion gap, generally in the range of 30 to

50 mEq/L. In such an instance, the acidosis is probably caused by gastrointestinal losses rather than by a renal lesion. The causes of acidosis characterized by a wide anion gap are listed in Table 28-10. In severe renal failure, inorganic compounds such as phosphates and sulfates are the major contributors to the increased anion gap. Organic compounds also accumulate in patients with severe renal failure. Ketoacidosis results from accelerated lipolysis and ketogenesis caused by relative or absolute insulin deficiency. Alcoholic ketoacidosis and starvation ketoacidosis result from the suppression of endogenous insulin secretion caused by inadequate carbohydrate ingestion. In addition, in alcoholic ketoacidosis, insulin resistance contributes to ketone formation. The syndrome of lactic acidosis results from impaired cellular respiration. Lactate is produced from the reduction of pyruvate in muscle, red blood cells, and other tissues as a consequence of anaerobic glycolysis. In situations of diminished oxidative metabolism, excess lactic acid is produced. This anaerobic state also favors a shift of keto acids to the reduced form, β-hydroxybutyrate. The nitroprusside reaction, which is catalyzed by the keto acids acetoacetate and acetone, is thus nonreactive in the setting of lactic acidosis. Lactic acidosis occurs most commonly in disorders characterized by inadequate oxygen delivery to tissues, such as shock, septicemia, and profound hypoxemia. Certain toxins may also sufficiently alter mitochondrial function and establish an effective anaerobic state. Some of these toxins may undergo metabolism into organic acids that can contribute to the generation of acidosis characterized by a large anion gap. Methanol is metabolized by alcohol dehydrogenase to formic acid. Ethylene glycol is metabolized to glycolic and oxalic acids. Salicylates are themselves acidic compounds and can cause acidosis characterized by a wide anion gap. The treatment of metabolic acidosis depends on the underlying cause and the severity of the manifestations. The rapid administration of parenteral sodium bicarbonate is generally indicated when the pH is less than 7.1 and hemodynamic instability is evident. Oral bicarbonate supplementation may be sufficient if the acidosis is caused by gastrointestinal bicarbonate loss or renal tubular acidosis (RTA). Treatment of organic acidosis should be directed at the underlying disorder. If the generation of the organic acid can be interrupted, the organic base pair may be metabolized, effectively regenerating bicarbonate. The acidemia of diabetic ketoacidosis, for example, can be effectively treated by administration of insulin, thereby inhibiting further ketogenesis. In lactic acidosis, therapy should be directed

Chapter 28—Fluid and Electrolyte Disorders toward improving tissue perfusion. In alcoholic and starvation ketoacidosis, administration of dextrose-containing intravenous fluids corrects the acidosis.

RENAL TUBULAR ACIDOSIS SYNDROMES Currently, three major RTA syndromes have been identified. These syndromes are discussed later, and their principal characteristics are described in Table 28-11.

Proximal Renal Tubular Acidosis Syndromes Proximal RTA occurs either alone or as the full Fanconi syndrome, with glycosuria, aminoaciduria, and phosphaturia. In proximal RTA, the threshold is reduced from 25 mmol/L to about 18 to 20 mmol/L (Fig. 28-9). Thus, a single-pulse loss of bicarbonate of about 850 to 900 mEq takes place.

Table 28-11 Renal Tubular Acidosis Syndromes Type

Locus

Defect

Proximal Hyperkalemic

S1-S3 CCD principal cell

Gradient limited

OMCD intercalated cells

↓ HCO3− threshold ↓ VM (−) leading to ↓ H+ secretion Three specific defects in H+ secretion

CCD, cortical collecting duct; OMCD, outer medullary collecting duct; S, segment; VM (−), negative transepithelial voltage in the OMCD.

319

Proximal RTA occurs in a significant number of systemic diseases, most notably Wilson disease, cystinosis, and the gammopathies, especially light-chain disease; it is also seen in renal transplantation. Several molecular defects have been identified that also cause proximal RTA: • Mutations in the carbonic anhydrase II gene, which reduce its expression. It is transmitted as an autosomal-recessive syndrome characterized by osteopetrosis, cerebral calcification, and mental retardation. • Mutations in the SCLC4A4b gene that codes for the basolateral membrane sodium-bicarbonate transporter. This lesion is also an autosomal-recessive disease characterized by glaucoma, cataracts, band keratopathy, and psychomotor retardation. • Mutations in the SLC9A3 gene, which encodes the luminal NHE3 transport protein. Generalized Fanconi syndrome, which appears to be the consequence of a deficit in adenosine triphosphate (ATP) production in the proximal tubule, which reduces the activity of the basolateral sodium-potassium adenosine triphosphatase (Na+,K+-ATPase). This reduction is seen in severe phosphate depletion as well as in hereditary fructose intolerance. The focus of treatment is to enhance proximal bicarbonate reabsorption by reducing ECF volume. This reduction is achieved most commonly by salt restriction. An attempt to raise serum bicarbonate by oral bicarbonate therapy is counterproductive because it raises extracellular volume, enhances bicarbonaturia, provokes kaliuresis and phosphaturia, and produces hypokalemia and hypophosphatemia. Supplemental potassium to correct the hypokalemia is often necessary.

8.5 8.0

Urine pH

7.5 7.0 6.5 6.0 5.5 5.0 4.5

11 12 13 15 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Serum bicarbonate mmol/L

Figure 28-9 Bicarbonate titration curve in proximal renal tubular acidosis (RTA). Normal bicarbonate titration is shown by the solid line. Note that in patients with proximal RTA, bicarbonate excretion begins to appear in the urine when serum bicarbonate concentration exceeds 16 to 18mmol/L, whereas in normal individuals, bicarbonate does not appear until serum bicarbonate is above 22 to 24mmol/L. Below the threshold, patients with proximal RTA can reabsorb filtered bicarbonate nearly completely.

320

Section VI—Renal Disease 8.5 8.0

Urine pH

7.5 7.0 6.5 6.0 5.5 5.0 4.5

11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Serum bicarbonate mmol/L

Figure 28-10 Bicarbonate titration curve in gradient-limit renal tubular acidosis. Note the fixed bicarbonate excretion at any level serum bicarbonate (compare with Fig. 28-9).

Hyperkalemic Renal Tubular Acidosis Syndromes This second major group of RTA is theoretically caused by defects in principal cell function and takes three clinical forms. Pseudohypoaldosteronism I is inherited as an autosomal-recessive disease characterized by hyperkalemia, sodium wasting, and failure to thrive. Serum aldosterone is elevated, and the defect is unresponsive to administration of mineralocorticoids. A truncated form of epithelial sodium channel (ENaC) is found with low activity and massive sodium loss and is treated by large amounts of oral sodium. A second form, termed familial hyperkalemic hypertension, or Gordon syndrome, is characterized by hyperkalemia and sodium avidity, often accompanied by low renin and mild hypertension. The defect is responsive to loop diuretics and sodium restriction. Gordon syndrome is also responsive to thiazide diuretics, which inhibit the sodium-chloride co-transporter (NCC) in the distal tubule. This form of hyperkalemic RTA is caused by a loss of function mutation in WNK-4 (with no lysine), which inhibits NCC function (discussed in Chapter 26). Thus, the activity of the NCC is increased, reducing delivery to the collecting duct and consequent reduction in flow and voltage mediated potassium excretion. Because sodium reabsorption is increased in the distal tubule, hypercalciuria is prominent, and nephrocalcinosis occurs. The third form of hyperkalemic RTA is an acquired disease usually associated with interstitial fibrosis. In this form, serum aldosterone and renin are reduced even in the presence of hyperkalemia. The reason for this defect in aldosterone secretion and responsiveness to administered mineralocorticoids is unknown but suggests defects in principal cell function. The disease responds partially to furo-

semide and cautious liberalization of salt intake without provoking hypertension. The third major category of RTA is the so-called gradientlimited form and appears to be caused by a defect in outer medullary collecting duct function. Both principal and intercalated cell transports are defective. This disease is characterized by a fixed defect in bicarbonate excretion without apparent threshold (Fig. 28-10) and a fixed excretion of alkaline urine. Most commonly, the disease is the consequence of an inherited defect in the activity of the hydrogenATPase or potassium-hydrogen-ATPase. Acquired damage to the latter can also occur in autoimmune diseases, most notably Sjögren syndrome. This damage also occurs in sickle cell disease and primary hyperparathyroidism when these are associated with interstitial kidney damage. Drugs such as amphotericin and lithium may also cause damage. Impaired function of the basolateral band III chloride-bicarbonate exchanger is also thought to explain isolated cases because mutations have been found in the AEI gene that codes for it. Multiple mutations in the AEI gene may also occur, and in such cases, RTA is associated with ovalocytosis. In all cases of distal, gradient-limited RTA, hyperchlor emic acidosis occurs and is accompanied, because of sodium loss, by secondary hyperaldosteronism, leading to potassium depletion. In children, the syndrome impairs growth, and it may be associated with hypokalemic muscle paralysis, hypercalciuria, and nephrocalcinosis. Therapy consists of bicar bonate replacement as well as potassium replacement. Particularly in children, large amounts of bicarbonate may be required to ensure normal growth. Distal RTA is also complicated by a low urinary excretion of citrate, which leads to severe nephrocalcinosis.

METABOLIC ALKALOSIS A gain in base or loss of acid increases the bicarbonate concentration of the ECF. Normally, an elevation of the serum

Chapter 28—Fluid and Electrolyte Disorders bicarbonate concentration is corrected by excretion of the excess bicarbonate. The maintenance of metabolic alkalosis, therefore, implies a defect in the renal mechanism regulating bicarbonate excretion. This failure to excrete excess bicarbonate occurs by both physiologic responses to volume depletion, especially if hypercapnia and hypokalemia accompany the alkalosis, or by pathophysiologic responses, as occur in autonomous mineralocorticoid excess. The most common cause of metabolic alkalosis is gastric loss of hydrochloric acid by vomiting or mechanical drainage. Diuretic (thiazide and loop) use is commonly associated with metabolic alkalosis. Volume depletion associated with vomiting and diuretic use enhances proximal bicarbonate reabsorption. Enhanced activity of sodiumhydrogen exchange at this site, and consequent enhanced volume reabsorption, results in enhanced bicarbonate reabsorption (see Chapter 26). Volume depletion also leads to aldosterone secretion, which stimulates distal nephron hydrogen secretion and augments potassium secretion. Repair of the alkalosis under these circumstances requires administration of sodium chloride and potassium. Endogenous or exogenous mineralocorticoid excess (see Fig. 28-7) is unresponsive to volume administration as extracellular volume is expanded. The stimulation of distal hydrogen secretion by aldosterone is sufficient to limit bicarbonate excretion and stimulate potassium secretion. Repair of this disorder requires removal of the excess mineralocorticoid. In all these disorders, concomitant hypokalemia promotes the maintenance of metabolic alkalosis. Excessive alkali ingestion (e.g., milk-alkali syndrome) is an uncommon cause of metabolic alkalosis and results from impaired renal bicarbonate excretion caused by renal failure in the setting of excess alkali intake. In this instance, both hypercalcemia and vitamin D excess are thought to play roles in damaging the kidney. Removal of alkali often corrects the alkalosis, but renal function remains reduced if nephrocalcinosis is prominent. The determination of urinary chloride concentrations is helpful in formulating a rational approach to the diagnosis and treatment of metabolic alkalosis. In patients with hypertension since childhood, alkalosis, hypokalemia, and low urinary chloride, consideration should be given to Liddle syndrome. These features resemble mineralocorticoid excess, but renin and aldosterone levels are suppressed (pseudohypoaldosteronism). The disorder is inherited as an autosomal recessive disorder with mutations in the ENaC gene that result in deletion of the C-terminal region of the protein. This condition results in reduced degradation and increased density of sodium channels in the luminal membrane of the principal cells of the collecting duct (see Chapter 26). The clinical features are a consequence of the enhanced salt reab sorption, resultant volume expansion, and increased distal nephron potassium and proton secretion. Patients with high urinary chloride and alkalosis and who are hypertensive require work-up for hypercorticism, which may be autonomous, as in primary aldosteronism and Cushing disease or secondary to renal artery stenosis. Rarer still are the 11-β-hydroxylase deficiencies, or apparent mineralocorticoid excess (AME), in which reduced conversion of glucocorticoids reaching the collecting duct leads to overstimulation of ENaC. Another such syndrome of congenital hypertension and alkalosis is glucocorticoid-remediable

321

aldosteronism (GRA), which is caused by a gene duplication in which the promoter for the 11-β-hydroxylase gene drives the aldosterone synthase gene and leads to adrenocorticotropic hormone–responsive aldosterone synthesis. In each of these instances, hyperactivity of ENaC leads to all clinical features of the syndrome. Normotensive or hypotensive conditions with alkalosis, hypokalemia, and high urinary chloride consist of two distinct forms: Bartter syndrome and Gitelman syndrome. Each syndrome involves distinct abnormalities in segmentspecific sodium chloride reabsorption, as well as differences in calcium and magnesium excretion. In Bartter syndrome (see Table 28-6), several disabling mutations in genes affecting the reabsorption of sodium chloride across the TAL have been characterized, including loss-of-function mutations in the NKCC2 co-transporter, the ROMK channel protein, and the basolateral chloride channel, and a gain in function mutation of the calcium-sensing receptor. Each of these mutations causes salt wasting, including enhanced calcium excretion, volume depletion, and, in many instances, reduced blood pressure. The reduction in ECF volume causes a secondary hyperaldosteronism, which, when coupled with enhanced sodium delivery to the collecting duct, causes potassium wasting and enhanced proton excretion. In Gitelman syndrome, disabling mutations in the DCT thiazide-sensitive sodium-chloride co-transporter have been described. The phenotypic characteristics of Gitelman syndrome in contradistinction to Bartter syndrome are the reduced calcium excretion and hypercalcemia observed, as might be expected from inhibition of the sodium-chloride co-transporter (see Chapter 26). The cause of hypermagnesuria in Gitelman syndrome is reduced expression of TRPM6 channel in the DCT.

RESPIRATORY ACIDOSIS Respiratory acidosis occurs with any impairment in the rate of alveolar ventilation. Acute respiratory acidosis occurs with a sudden depression of the medullary respiratory center (narcotic overdose), with paralysis of the respiratory muscles, and with airway obstruction. Chronic respiratory acidosis generally occurs in patients with chronic airway disease (emphysema), with extreme kyphoscoliosis, and with extreme obesity (pickwickian syndrome). The serum bicarbonate concentration is increased, the magnitude of which depends on the acuity and the severity of the respiratory disorder. The compensatory increase in serum bicarbonate in prolonged hypercapnia (>1 week) is primarily a function of bone buffering as the kidney plays a relatively minor role in the increase. Acute increases in the Pco2 result in somnolence, confusion, and, ultimately, carbon dioxide narcosis. Asterixis may be present. Because carbon dioxide is a cerebral vasodilator, the blood vessels in the optic fundi are often dilated, engorged, and tortuous. Frank papilledema may be present in patients with severe hypercapnic states. The only practical therapy of acute respiratory acidosis involves treatment of the underlying disorder and ventilatory support. In patients with chronic hypercapnia who develop an acute increase in the Pco2, attention should be directed toward identifying the factors that may have aggravated the chronic disorder. Diuretics often exacerbate the

322

Section VI—Renal Disease

increase in serum bicarbonate and result in a mixed disorder of metabolic alkalosis and respiratory acidosis. Under these circumstances, acidification of the serum may be necessary to improve ventilation.

RESPIRATORY ALKALOSIS Respiratory alkalosis occurs when hyperventilation reduces the arterial Pco2 and consequently increases the arterial pH. Acute respiratory alkalosis is most commonly a result of pregnancy; it may also occur in damage to the respiratory centers, in acute salicylism, in fever and septic states,

in advanced liver disease, and when respiratory rate is increased in pneumonia, pulmonary embolism, and congestive heart failure. The disorder may be produced iatrogenically by injudicious mechanical ventilatory support. Chronic hyperventilation occurs in the acclimatization response to high altitudes cause by reduced ambient partial pressure of oxygen. Acute hyperventilation is characterized by lightheadedness, paresthesias, circumoral numbness, and tingling of the extremities. Tetany occurs in severe cases. When anxiety provokes hyperventilation, air rebreathing with a paper bag generally terminates the acute attack.

Prospectus for the Future • Identification of additional gene mutations informative about the renal regulation of water and solute transport

References Andreoli TE: Water: Normal balance, hyponatremia and hypernatremia. Ren Fail 22:711-735, 2000. Cao G, Hoenderop JGJ, Bibdels RJM: Insight into the molecular regulation of the epithelial magnesium channel TRPM6. Curr Opin Nephrol Hypertens 17:373-378, 2008. Gamba G: Role of WNK kinases in regulating tubular salt and potassium transport and in the development of hypertension. Am J Physiol Renal Physiol 288:F245F252, 2005.

• Design of small molecules to inhibit salt and water transport in specific nephron segments for the treatment of hypertension and edema

Ellison DH, Berl T: Clinical practice: The syndrome of inappropriate antidiuresis. N Engl J Med 356:2064-2072, 2007. Kokko JP: Fluid and electrolytes. In Goldman L, Bennett JC (eds): Cecil Textbook of Medicine, 21st ed. Philadelphia, WB Saunders, 2000, pp 540-567. Tannen RL: Dyskalemias. In Massry SG, Glasscock FJ (eds): Textbook of Nephrology, 4th ed. Philadelphia, Lippincott Williams & Wilkins, 2001, pp 295-307.

Chapter

Glomerular Diseases Jamie P. Dwyer and Julia B. Lewis

ach human kidney contains nearly 1 million glomerular capillary tufts, which derive from an afferent arteriole and are held together by the mesangial matrix. The glomerular capillary tufts drain into an efferent arteriole forming an arteriolar portal system. Fenestrated endothelial cells, the glomerular basement membrane (GBM), and delicate foot processes extending from epithelial podocytes, which are interconnected to each other by slit diaphragms (Web Fig 29-1), form a selective filtration barrier between the capillary blood and urinary space (Web Fig 29-2A). Each glomerular tuft has an associated renal tubule that drains into the urologic system. Nearly one fourth of the blood of each heartbeat is filtered by the kidney (about 120 to 180L per 24 hours). Remarkably, despite the 12,000 to 18,000g of protein filtered by the capillaries each day, less than 150mg appears in urine. This is accomplished in part by the negatively charged GBM and the 4-nm slit-pore membranes restricting the movement of large or negatively charged proteins. Therefore, the glomerulus serves as a size- and chargeselective barrier to the movement of proteins and cells from the capillary blood into the urinary space. The glomerulus can be injured by a variety of means, including genetic mutations producing familial diseases, immune-mediated inflammation, vascular injury, deposition of abnormal proteins, and infection. These injuries result in diverse glomerular diseases manifesting clinically as proteinuria, hematuria, pyuria, and vascular changes.

GFR. If the loss of GFR (seen as a rise in serum creatinine) occurs over days, acute nephritis is called rapidly progressive glomerulonephritis (RPGN) and is associated with crescentic glomerulonephritis on renal biopsy. Patients with the clinical syndrome of RPGN in whom the disease extends to the lungs are classified as having a pulmonary-renal syndrome (Table 29-2). In nephrotic syndrome, the patient excretes more than 3.5g of protein in a 24-hour urine collection and has edema, hypoalbuminemia, and hypercholesterolemia. In long-standing nephrotic syndrome, decreased GFR and hypertension often develop (e.g., in diabetic nephropathy). Many glomerular diseases present with microscopic or gross hematuria and either no or mild proteinuria. Hematuria may be the only manifestation of some glomerular diseases throughout their course, as in thin basement membrane disease, or may be an early manifestation that progresses over time to involve other clinical signs such as decreased GFR, as in antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis. When abnormal proteins, such as paraproteins, are deposited in or accumulate in the glomerulus in glomerular deposition diseases, the clinical manifestations can range from asymptomatic mild proteinuria to severe nephrotic syndrome. Glomerular vascular syndromes occur in patients in whom the injury is primarily

Clinical Syndromes Glomerular diseases can have diverse clinical manifestations, including hematuria (Web Fig. 29-3), proteinuria, pyuria (Web Fig. 29-4), hypertension, fluid retention, edema, and a reduction in glomerular filtration rate (GFR). Glomerular diseases can be acute, developing over days; subacute, developing over weeks; or chronic, developing over months or years. Distinct clinical syndromes have been described; however, these syndromes are not always mutually exclusive (Table 29-1) The acute nephritic syndrome is characterized by hypertension, hematuria, edema, red blood cell (RBC) casts (Fig. 29-1) or dysmorphic RBCs (Web Fig. 29-5), modest proteinuria (1 to 2 g per 24 hours), and decreased

Figure 29-1 Immunofluorescence demonstrating a linear pattern of immunoglobulin G.

323

324

Section VI—Renal Disease

Table 29-1 Clinical Syndromes Acute Nephritic Syndromes Poststreptococcal glomerulonephritis Subacute bacterial endocarditis Lupus nephritis* (WHO class III or IV) Anti-GBM disease, Goodpasture syndrome ANCA-associated vasculitis (Wegener granulomatosis, microscopic polyangiitis, Churg-Strauss syndrome)* Cryoglobulinemia* IgA nephropathy Henoch-Schönlein purpura Membranoproliferative glomerulonephritis Nephrotic Syndromes Minimal change disease FSGS MGN Diabetic nephropathy Lupus nephritis (WHO class V) ANCA-associated vasculitis Deposition diseases Nail-patella syndrome Fabry disease Syphilis (MGN) Schistosomiasis (MPGN, FSGS, amyloid) Primarily Hematuria Manifested Glomerular Diseases IgA nephropathy Thin basement membrane disease Alport syndrome MGPN Lupus nephritis (WHO class II or III) ANCA-associated vasculitis (early)* Sickle cell disease Glomerular Deposition Diseases Light-chain deposition disease Amyloidosis Fibrillary glomerulonephritis, immunotactoid glomerulonephritis Fabry disease Glomerular Vascular Syndromes Hypertensive nephrosclerosis Cholesterol emboli Sickle cell disease Thrombotic thrombocytopenia purpura, hemolytic uremic syndrome Antiphospholipid antibody syndrome ANCA-associated vasculitis Henoch-Schönlein purpura Cryoglobulinemia* Amyloidosis Ischemic nephropathy Infection-Associated Syndromes Poststreptococcal glomerulonephritis Subacute bacterial endocarditis HIV (FSGS) Hepatitis B and C (MGN and MPGN, respectively) Syphilis Leprosy Malaria Schistosomiasis *May present as a pulmonary-renal syndrome. ANCA, antineutrophil cytoplasmic antibody; FSGS, focal segmental glomerulosclerosis; GBM, glomerular basement membrane; IgA, immunoglobulin A; MGN, membranous glomerulonephritis; MPGN, membranoproliferative glomerulonephritis; WHO, World Health Organization.

Table 29-2 Differential Diagnosis of Rapidly Progressive Glomerulonephritis Linear Immune Staining Anti-GBM disease Goodpasture syndrome Rarely membranous glomerulonephritis Granular Immune Staining Subacute bacterial endocarditis (past infectious) Lupus nephritis Cryoglobulinemia Membranoproliferative glomerulonephritis (type II more than type I) Immunoglobulin A nephropathy, Henoch-Schönlein purpura Idiopathic No Immune Staining (Pauci-immune) Antineutrophil cytoplasmic antibody–associated vasculitis (Wegener granulomatosis, microscopic polyangiitis, Churg-Strauss syndrome) Idiopathic

localized to the renal vasculature and is usually associated with hematuria and mild proteinuria. Lastly, a wide variety of infections can produce inflammatory reactions in the glomerulus, ranging from nephritic syndrome with RPGN to mild proteinuria or nephrotic syndrome. The classification of glomerular diseases is hampered by the fact that an individual glomerular disease can present with more than one constellation of clinical signs or symptoms. For example, lupus nephritis can present as nephrotic syndrome, rapidly progressive glomerulonephritis, or asymptomatic hematuria. Hence, all classifications of glomerular diseases are complex and somewhat arbitrary. Each of the major glomerular diseases is discussed next, and potential alternate manifestations are noted (see Table 29-1).

Evaluation of Glomerular Diseases A detailed history and physical examination can help clarify the differential diagnosis of glomerular lesions. Onset and timing may be important (e.g., nephrotic syndrome in a child suggests minimal change nephropathy). Associated physical examination findings may add to the diagnosis (e.g., Raynaud phenomenon in lupus nephritis, livedo reticularis in cholesterol emboli). Assessment for anemia, thrombocytopenia, eosinophilia, microangiopathic hemolysis, serologies for antinuclear antibody (ANA), ANCAs, anti-GBM antibody, antibodies to hepatitis B and C, HIV, rheumatoid factor, anti-DNAse B, antistreptolysin O (ASO) titer, cryo globulin, and monoclonal proteins may help in narrowing the diagnostic possibilities. Urine microscopy is a critical element in the evaluation of a patient with glomerular diseases and may reveal hematuria, RBC casts, oval fat bodies or fatty casts (Web Fig. 29-6), or proteinuria. Measurement of 24-hour creatinine clearance and proteinuria helps distinguish the presence or absence of nephrotic syndrome and assesses GFR. Ultimately, a specific diagnosis may rely on a renal biopsy. Indications for renal biopsy vary from patient to patient and between countries. Renal biopsy allows one

Chapter 29—Glomerular Diseases

325

Table 29-3 Glomerular Diseases Associated with Hypocomplementemia Poststreptococcal glomerulonephritis Lupus nephritis (acute) Cholesterol emboli Membranoproliferative glomerulonephritis, hepatitis C Subacute bacterial endocarditis Cryoglobulinemia Shunt nephritis

Acute Nephritic Syndromes Figure 29-2 Glomerulus demonstrating crescent formation.

to make a diagnosis and can be determined by light microscopy if the disease is focal (involving 50% of glomeruli), segmental (involving a portion of a glomerulus) or global (involving the entire glomerulus). It can also demonstrate so-called crescent formation (Fig. 29-2), which is the pathologic hallmark of rapidly progressive glomerulonephritis. Immunofluorescence microscopy can identify types of immunoglobulin deposition and location (e.g., linear immunoglobulin G [IgG] staining in Goodpasture syndrome (Fig. 29-1). There may be minimal or no immune deposition (pauci-immune glomerulonephritis). Electron microscopy can show the location of electron-dense deposits (e.g., subepithelial deposits in poststreptococcal glomerulonephritis (Web Fig. 29-7) or subendothelial deposits in proliferative lupus nephritis (Web Fig. 29-8).

General Treatment Guidelines for Glomerular Disease Specific therapies for the different glomerular disease are noted next. In all forms of glomerular disease, treatment of hypertension, if present, is indicated with a target blood pressure of 130/80mmHg. For many glomerular diseases, including all those associated with proteinuria, treatment with drugs that inhibit the renin-angiotensin system is recommended. It has also been suggested that accelerating therapy to maximally reduce proteinuria should be a goal. Volume overload, manifested by edema, should be treated by reduction in salt and water intake and judicious use of diuretics. Hypercholesterolemia should be controlled with dietary modification and pharmacologic therapy. If a hypercoagulable state exists, anticoagulation may be necessary. Every effort should be made to avoid exposure to nephrotoxins because patients with glomerular disease may be at increased risk for acute kidney injury. Lastly, close monitoring of renal function by a specialist is indicated in most cases.

POSTSTREPTOCOCCAL GLOMERULONEPHRITIS Poststreptococcal glomerulonephritis (PSGN) can occur as a postinfectious complication of skin and throat infections with particular M types (nephritogenic strains) of streptococci. PSGN due to streptococcal pharyngitis occurs in fewer than 5% of people infected, typically 1 to 3 weeks after the pharyngitis. Streptococcal impetigo is a less common cause of PSGN than pharyngitis but leads to PSGN in as many as 50% of infected people 2 to 6 weeks after the impetigo. In undeveloped countries, PSGN can occur in a epidemic form, but in Western countries, it typically occurs sporadically in the summer and autumn. PSGN can occur in adults but usually occurs in children between the ages of 2 and 14 years. Patients present classically with acute nephritis, characterized by hematuria, pyuria, RBC casts, edema, hypertension, systemic symptoms of headache and malaise, flank pain due to renal capsular swelling, and oliguric renal failure. Because the hematuria occurs after the pharyngitis, it is called metapharyngitic or postpharyngitic hematuria. Five percent of children and 20% of adults have nephrotic range proteinuria. A subclinical disease has also been reported, characterized by asymptomatic microscopic hematuria. Early in the course, 90% of patients will have decreased levels of C3 and CH50 with normal levels of C4. Patients with PSGN must be distinguished from other glomerular diseases associated with hypocomplementemia (Table 29-3). Rheumatoid factor, cryoglobulins, and ANCA may all be positive. Increased titers of ASO antibodies (30%), anti-DNAase (70%) or antihyaluronidase antibodies (40%) can help confirm the diagnosis. Histologically, the kidney demonstrates a diffuse proliferative glomerulonephritis with hypercellularity of mesangial and endothelial cells (Web Figs. 29-9 and 29-10), glomerular infiltrates of polymorphonuclear leukocytes, and granular, “lumpy-bumpy” subendothelial and subepithelial deposits of IgG, IgM, and complement. Treatment is supportive and may require renal replacement therapy. Antibiotic therapy does not alter the course of PSGN. Immunosuppressive therapy is also ineffective. Complete recovery occurs in 90% to 95% of patients; in children, recovery is usually seen within 3 to 6 weeks of the onset of nephritis, whereas in adults, proteinuria and hematuria may continue for 1 to 2 years. End-stage renal disease (ESRD) is uncommon, occurring in 1% to 3% of adults and rarely in children.

326

Section VI—Renal Disease

SUBACUTE BACTERIAL ENDOCARDITIS Endocarditis-associated glomerulonephritis is a complication of subacute bacterial endocarditis. Patients present with hematuria, pyuria, and mild proteinuria, or less commonly with RPGN. Laboratory examination reveals an elevated erythrocyte sedimentation rate, hypocomplementemia, a positive rheumatoid factor, cryoglobulinemia (type III), and anemia. Renal biopsy reveals a focal proliferative glomerulonephritis with abundant mesangial, subendothelial, and subepithelial deposits of IgG, IgM, and complement. Patients who present with RPGN have crescents on renal biopsy. Embolic infarcts or abscesses may also be present. Treatment is antibiotics for 4 to 6 weeks, and the prognosis is good. Glomerulonephritis can also occur in patients with infected ventriculoatrial and ventriculoperitoneal shunts (shunt nephritis); pulmonary, intra-abdominal, pelvic, or cutaneous infections; and infected vascular prostheses. Treatment is eradication of the infection.

LUPUS NEPHRITIS Sixty percent of adults and 80% of children with systemic lupus erythematosus (SLE) develop renal abnormalities. The clinical manifestations and treatment of lupus nephritis are closely linked to the renal pathology (Table 29-4). Clinical signs and laboratory data can include hematuria, proteinuria, RBC casts, hypertension, hypocomplementemia, and anti–double-stranded DNA (anti-dsDNA) antibodies. Renal biopsy is critical to distinguish the variants of lupus nephritis (Web Fig. 29-11), and patients with lupus often undergo multiple biopsies as their lupus-related renal lesions may vary over time. The World Health Organization (WHO) has outlined distinct patterns of lupus-related glomerular injury (see Table 29-4). WHO class I lesions have minimal clinical manifestations and either normal histology or minimal mesangial deposits. Prognosis is excellent, and no treatment is required. WHO class II nephritis demonstrates mesangial immune complex deposition with mesangial proliferation, but few clinical renal manifestations, normal renal function, and a good prognosis. Specific treatment is generally not necessary. WHO class III lesions demonstrate focal lesions

with proliferation and scarring (focal proliferative lupus nephritis), and patients with class III lesions have diverse clinical courses. Patients can present with hypertension, hematuria, proteinuria, nephrotic syndrome (25% to 33%), or elevated serum creatinine (25%). Some patients with only mild focal proliferation involving a small percentage of glomeruli respond to steroid therapy alone with an excellent prognosis. Others with more severe proliferation involving a greater percentage of glomeruli have a worse prognosis, and therapy with steroids and other immunosuppressive drugs (cyclophosphamide, mycophenolate) is required. WHO class IV nephritis demonstrates global, diffuse proliferative lesions involving most of the glomeruli (diffuse proliferative lupus nephritis). Clinically, patients present with the most severe manifestations, including RPGN, hematuria, RBC casts, hypertension, proteinuria (50% nephrotic range), and declining renal function. Without treatment, WHO class IV lesions have the worst prognosis, and therapy with steroids and immunosuppressive drugs is recommended. If remission is achieved (defined as a return to near-normal renal function and proteinuria 10 years) type 1 diabetes who have other complications, such as peripheral and autonomic neuro pathy, nephropathy, and retinopathy, GI complaints are also common within the first decade of diagnosis. Diabetic gastroparesis appears to occur as a result of permanent neuropathy of autonomic and enteric nerves, transitory variations in glycemic control, or a combination of both. Idiopathic gastroparesis is also common and comprises those instances with no clearly identifiable cause. Up to one third of these patients have virus-induced gastroparesis, with viral infiltration of the myenteric plexus in the stomach. Patients who have undergone gastric surgery, especially those having had preoperative gastric outlet obstruction as a complication of PUD, are also commonly affected by

Table 37-4 Causes of Delayed Gastric Emptying Mechanical Causes Peptic ulcer disease, scarred pylorus Malignancy: gastric cancer, gastric lymphoma, pancreatic cancer Gastric surgery: vagotomy, gastric resection, roux-en-Y anastomosis Crohn disease Endocrine and Metabolic Causes Diabetes mellitus Hypothyroidism Hypoadrenal states Electrolyte abnormalities Chronic renal failure Medications Anticholinergics Opiates Dopamine agonists Tricyclic antidepressants Abnormalities of Gastric Smooth Muscle Scleroderma Polymyositis, dermatomyositis Amyloidosis Pseudo-obstruction Myotonic dystrophy Neuropathy Scleroderma Amyloidosis Autonomic neuropathy Central Nervous System or Psychiatric Disorders Brainstem tumors Spinal cord injury Anorexia nervosa Stress Miscellaneous Idiopathic gastroparesis Gastroesophageal reflux disease Nonulcer (functional) dyspepsia Cancer cachexia or anorexia

428

Section VII—Gastrointestinal Disease

gastroparesis. Finally, Parkinson disease, rheumatologic disorders, hypothyroidism or hyperthyroidism, chronic intestinal pseudo-obstruction, and a variety of paraneoplastic syndromes can also produce gastroparesis. The diagnostic evaluation of delayed gastric emptying should focus on excluding structural and metabolic abnormalities. Endoscopy is the preferred initial test to rule out mechanical gastric outlet obstruction, and a small bowel follow-through radiograph may be useful to exclude small bowel lesions. Serum electrolytes, blood cell counts, and thyroid studies should also be performed. When these studies are negative, radionuclide scintigraphy (gastricemptying scan) using a mixed solid-liquid meal can quantitate delayed gastric emptying. Assessment of solid emptying is more clinically relevant than liquid emptying. In especially difficult cases, GI manometry and electrogastrography may help in the diagnosis. Managing gastroparesis begins with identifying and treating potentially correctable causes. Medications that reduce gastric emptying, such as narcotics, anticholinergics, and tricyclic antidepressants, should be avoided. Because liquids empty easier than solids, and because liquid emptying is often preserved in patients with gastroparesis, simple dietary modifications may be helpful in treatment. The diet should be modified to include blenderized foods and liquid supplements. High-fat and fiber-rich foods should be avoided because they inhibit gastric emptying under normal conditions and are less likely to empty. Medical options are limited and involve the use of prokinetic drugs, which are agents that improve transit in the GI tract. Metoclopramide is a dopamine-2 receptor antagonist that also facilitates the release of acetylcholine from cholinergic nerve terminals in the gut, thereby accelerating gastric emptying. The efficacy of metoclopramide is inconsistent, and adverse effects and the development of tolerance complicate long-term therapy. Adverse effects occur in up to 20% of patients and include drowsiness, anxiety, fatigue, insomnia, restlessness, agitation, extrapyramidal effects, galactorrhea, and menstrual irregularities. The typical dosage is 10mg, 20 to 30 minutes before meals and at bedtime, although doses as high as 80mg or as low as 20mg may be used daily. Doses should be reduced for patients with renal failure. Domperidone, another dopamine receptor antagonist with prokinetic properties, has similar efficacy to metoclopramide in the treatment of delayed gastric emptying but is currently not available in the United States. Cisapride, an agent that increases gastric motor activity by facilitating the release of acetylcholine at the myenteric plexus, is no longer routinely available in the United States and other countries because of serious adverse effects, including ventricular tachycardia, ventricular fibrillation, torsades de pointes, and prolongation of the QT interval, which have been reported when cisapride is administered with other drugs that inhibit cytochrome P-450. Erythromycin is a macrolide antibiotic that stimulates smooth muscle motilin receptors located at all levels of the GI tract. The prokinetic effects of erythromycin are related to its ability to mimic the effect of the GI peptide motilin to stimulate smooth muscle contraction, which accounts for the acceleration of solid and liquid gastric emptying. Erythromycin may dramatically improve gastric emptying in patients with severe diabetic gastroparesis when given acutely

at an intravenous dose of 1 to 3mg/kg every 8 hours. Longterm use of the drug at a dose of 250 to 500mg orally every 8 hours in patients with gastric stasis is of limited efficacy because of tachyphylaxis and side effects. Endoscopic botulinum toxin A injection into the pyloric sphincter has also been reported in the treatment of delayed gastric emptying in small studies, but long-term benefit has not been proved. In patients who are refractory to these measures, surgical placement of a jejunal tube, with or without a venting gastrostomy, may be necessary. Total parenteral nutrition is rarely indicated. Surgical gastrectomy should only be considered in patients with refractory postsurgical gastric stasis. Gastric pacemakers and other prokinetics, specifically new serotonin-receptor agonists, are under investigation and may be options in the future.

Rapid Gastric Emptying Rapid gastric emptying is a far less common clinical problem than delayed gastric emptying. Dumping syndrome describes the alimentary and systemic manifestations of early delivery of large amounts of osmotically active food to the small intestine. Dumping syndrome is usually seen when the normal reservoir, grinding, and sieving properties of the stomach are disrupted, most commonly following surgery for obesity (Roux-en-Y gastric bypass) or PUD. The accelerated emptying of hypertonic boluses of nutrient material into the small intestine results in splanchnic vasodilation and release of vasoactive peptides. Early dumping symptoms, occurring about 30 minutes after a meal, include epigastric fullness and pain, nausea, vomiting, early satiety, and vasomotor features such as flushing, palpitations, and diaphoresis. Later symptoms, such as diaphoresis, tremulousness, and weakness, occur about 2 hours after a meal and may be caused by hypoglycemia from rebound hyperinsulinemia. Treatment of dumping syndrome involves dietary manipulation to decrease the volume and osmotic load emptied into the intestine. Frequent small feedings of meals low in carbohydrates, separation of liquid and solid intake, and avoidance of hypertonic fluids and lactose are usually helpful. When these measures fail, administration of octreotide at a dose of 25 to 50 mcg subcutaneously 30 minutes before meals may be helpful. Octreotide acts by slowing gastric emptying and intestinal transit as well as by inhibiting the release of insulin. Surgical procedures to slow gastric emptying have limited success.

Gastric Volvulus Gastric volvulus occurs when the stomach twists on itself. This event may be transient, producing few if any symptoms, or may lead to obstruction or even ischemia and necrosis. Primary gastric volvulus, seen in one third of the patients, occurs below the diaphragm when the stabilizing ligaments are too lax as a result of congenital or acquired causes. Secondary gastric volvulus occurs above the diaphragm in association with paraesophageal hernias or other diaphragmatic defects. Acute gastric volvulus produces sudden, severe pain of the upper abdomen or chest, persistent retching

Chapter 37—Diseases of the Stomach and Duodenum producing scant vomitus, and the inability to pass a nasogastric tube. This combination of symptoms, also known as Borchardt triad, should lead to a strong clinical suggestion of acute gastric volvulus. Chronic gastric volvulus may be associated with mild and nonspecific symptoms, such as epigastric discomfort, heartburn, abdominal fullness or bloating, and borborygmi, especially after meals. The diag-

429

nosis of gastric volvulus is made by upper GI series demonstrating an abrupt obstruction at the site of the volvulus. Acute gastric volvulus requires emergency surgical evaluation because of the substantial risk for mortality related to gastric ischemia or perforation. Treatment consists of surgical gastropexy and repair of any associated paraesophageal hernia.

Prospectus for the Future Our understanding of a significant number of issues involving gastroduodenal pathologic mechanisms and therapeutics will continue to evolve in the next few years. Select goals include the following: • Development of variations in the structure and delivery of antisecretory therapy, leading to formulations designed to provide increased rapid onset of action and to improve effectiveness in individuals who are unable to take medications by mouth or those with dysmotility or malabsorption • Further clarification of the optimal H. pylori treatment regimen given increased antibiotic resistance profiles

References Chan FK, Graham DY: NSAIDs, risks, and gastroprotective strategies: Current status and future. Gastroenterology 134:1240-1257, 2008. Fuccio L, Minardi ME, Rocco MZ, et al: Meta-analysis: Duration of first-line proton-pump inhibitor-based triple therapy for Helicobacter pylori eradication. Ann Intern Med 147:553-562, 2007. Olsen KM, Devlin JW: Comparison of the enteral and intravenous lansoprazole pharmacodynamic responses in critically ill patients. Aliment Pharmacol Ther 28:326-333, 2008.

and the goal of improving patient compliance to treatment • Procurement of evidence-based data to aid in managing stress-related mucosal ulcerations in critically ill patients • Elucidation of the exact nature and strength of the association between COX-2–selective inhibitors and cardiovascular disease and future development of safer NSAIDs • Further insight into the mechanism governing GI motility and the development of new agents for the treatment of motility disorders

Papatherodoridis GV, Sougioultzia S, Archimandritis AJ: Effects of Helicobacter pylori and nonsteroidal anti-inflammatory drugs on peptic ulcer disease: A systematic review. Clin Gastroenterol Hepatol 4:130-142, 2006. Park MI, Camilleri M: Gastroparesis: clinical update. Am J Gastroeterol 101:11291139, 2006. Vergara M, Catalan M, Gisbert JP, et al: Meta-analysis: Role of Helicobacter pylori eradication in the prevention of peptic ulcer in NSAID users. Aliment Pharmacol Ther 21:1411-1418, 2005.

VII

Chapter

Inflammatory Bowel Disease Christopher S. Huang and Francis A. Farraye

lthough a significant number of infectious organisms and noninfectious processes (e.g., medications, radiation, ischemia) can result in intestinal inflammation, the term inflammatory bowel disease (IBD) generally refers primarily to two idiopathic diseases: ulcerative colitis and Crohn disease. The diagnosis of IBD is made by incorporating clinical, endoscopic, radiologic, and histologic information. Ulcerative colitis is characterized by inflammatory changes that involve the colonic mucosa in a continuous superficial fashion, generally starting in the rectum and extending proximally. Depending on the extent of the disease, ulcerative colitis can be divided into proctitis (rectum only), proctosigmoiditis, left-sided colitis (extending to the splenic flexure), or pancolitis. This classification is important for both prognostic and therapeutic reasons. Unlike ulcerative colitis, Crohn disease can involve any segment of the gastrointestinal system, often in a discontinuous fashion. It is characterized by transmural inflammation, which results in significant complications such as abscesses, fistulas, and strictures. Despite the chronic nature of these two diseases, new and emerging targeted anti-inflammatory treatments hold great promise in helping to reduce morbidity and improve the quality of life of individuals with IBD.

Epidemiology In the United States, about 1.4 million individuals have IBD, and the overall incidence of new cases of IBD is about 3 to 10 new cases per 100,000 people. During the past several decades, the incidence of ulcerative colitis has remained stable, whereas the incidence of Crohn disease has gradually increased. The prevalence of IBD is essentially 10-fold higher, between 30 and 100 per 100,000 people. A bimodal age of presentation exists, with an initial peak between the second and fourth decades of life followed by another peak at about the sixth decade of life. Both sexes are equally affected. The incidence and prevalence of IBD reflect the genetic and environmental factors that contribute to these disorders. 430

For example, both diseases are common in northern climates, among whites, particularly in populations with Northern European ancestry, including North Americans, South Africans, and Australians. Individuals of Ashkenazi Jewish descent have also been found to have a twofold to eightfold increased risk for these disorders compared with non-Jews. Although the prevalence and incidence rates of IBD are lowest in Hispanics and Asians, IBD can occur in any ethnic or racial group from anywhere in the world.

Causes Although the causes of IBD remain unknown, recent advances in the understanding of the genetic, immunologic, and environmental factors are beginning to decipher the etiologic factors of these complex disorders. Currently, it is believed that IBD results from an inappropriate, overactive mucosal immune response to commensal intestinal bacteria in genetically susceptible individuals.

GENETIC FACTORS About 5% to 20% of patients with IBD have a first-degree relative with the disease, and first-degree relatives of IBD patients have about a 10- to 15-fold increased risk for developing IBD, predominantly with the same disease as the proband. A positive family history is generally more frequently observed in patients with Crohn disease than in patients with ulcerative colitis, suggesting that genetic factors contribute more significantly in the etiology of Crohn disease. Through advances in genome-wide association studies, several susceptibility loci on multiple chromosomes have been linked to IBD, supporting a polygenic cause to these disorders. Polymorphisms in the NOD-2 gene (also known as CARD-15, located on chromosome 16) were the first definitive genetic risk factors identified for Crohn disease. Homozygous mutations of the NOD-2 gene are associated with a greater than 20-fold increase in susceptibility for Crohn disease. Defects in the NOD-2 protein appear

Chapter 38—Inflammatory Bowel Disease

431

to result in abnormal intestinal immune responses to bacterial cell wall components. These gene mutations are estimated to account for about 15% to 25% of the cases of Crohn disease and are linked predominantly to fibrostenotic terminal ileal disease. Another gene of interest is the IL-23R gene on chromosome 1, which has also been shown to be strongly associated with IBD, particularly Crohn disease. Disease-associated IL-23R polymorphisms have also been reported in patients with ulcerative colitis, psoriasis, and ankylosing spondylitis. Many other genes have been uncovered by genome-wide association studies, such as OCTN, DLG-5, ICAM-1, and TLR-4, and are currently under investigation for their potential involvement in the pathogenesis of IBD.

ized. It has been postulated that poor sanitation, food contamination, and crowded living conditions are associated with helminthic infection, which leads to regulatory T-cell conditioning and stimulation of IL-10 and transforming growth factor-β production by mononuclear cells, thereby preventing intestinal inflammation. However, to date, the only environmental factor clearly associated with IBD is tobacco smoking. Smoking seems to be protective against ulcerative colitis, whereas smokers with Crohn disease have more aggressive disease than do nonsmokers. No dietary triggers have been found to cause IBD, but elemental diets and diversion of the fecal stream can reduce recurrence of inflammation in Crohn disease.

IMMUNOLOGIC FACTORS

Clinical Features of Ulcerative Colitis

Profound alterations in mucosal immunology have been demonstrated in patients with IBD. In the normal immunologic state of the intestine, recently activated lymphoid tissue is abundant within the mucosal compartment. This state has been described as controlled, or physiologic, inflammation, which has likely developed in response to constant encounters with antigenic substances (derived from host microbial flora, or dietary and environmental sources) that have crossed the epithelial barrier from the luminal environment. Indeed, one of the main functions of the intestinal immune system is to discriminate noxious or harmful substances and organisms from nonharmful ones. As a result, a large and well-maintained network of many different mucosal immune cells exists, including cells involved in reducing immune responses (regulatory cells) and those involved in activating immune responses. In IBD, this homeostatic balance, or immune tolerance, is dysregulated, resulting in overactivation of the immune system. Crohn disease, for example, reflects an excessive and persistent CD4 helper T-cell subtype 1 (TH1) immune response to components of commensal bacterial flora. The TH1 cytokine profile, including interferon-γ, interleukin-2 (IL-2), IL-12, and tumor necrosis factor-α (TNF-α), is elevated in patients with Crohn disease. The cytokine profile of ulcerative colitis is atypical, with greater expression of IL-5 and IL-13 present, cytokines characteristically associated with a TH2 response. More recently, non-TH1/ TH2 pathways have been identified as being potentially important in the pathogenesis of IBD. IL-23, for example, has been recognized as an inducer of a subset of proinflammatory T cells (TH17) that secrete high levels of IL-17 and play an important role in mediating inflammation in murine models of colitis. IL-17 expression has been shown to be upregulated in active IBD, both Crohn disease and ulcerative colitis.

ENVIRONMENTAL FACTORS Although believed to be important, the role of environmental factors in IBD pathogenesis remains poorly understood. Many infectious agents, including Mycobacteria paratuberculosis and measles virus, have been implicated in IBD, but none fulfills the criteria of true pathogens. Environmental factors are suspected because the disease is more common in industrialized countries, and the frequency has been increasing in countries that are becoming more industrial-

Ulcerative colitis is characterized by chronic inflammation of the mucosal surface that involves the rectum (proctitis) and extends proximally through the colon in a continuous manner. The extent and severity of the colonic inflammation determine prognosis and presentation (insidious versus acute onset). Most patients initially exhibit diarrhea, abdominal pain, urgency to defecate, rectal bleeding, and the passage of mucus per rectum. Patients occasionally have extraintestinal manifestations (see later discussion) before they develop intestinal symptoms. About 40% to 50% of patients have proctitis or proctosigmoiditis, 30% to 40% have left-sided colitis (disease extending to the splenic flexure), and the remaining 20% to 25% have pancolitis. Of the patients who initially show proctitis or proctosigmoiditis, about 15% develop more extensive disease over time. The typical clinical course is of chronic intermittent exacerbations, followed by periods of remission. Signs of a worsening clinical course include the development of abdominal pain, dehydration, fever, and tachycardia. Clinical features, including bowel frequency, fever, increased heart rate, and blood in stools, as well as the presence of anemia and an elevated erythrocyte sedimentation rate (ESR) or C-reactive protein, have been used to assess severity of ulcerative colitis.

MAJOR COMPLICATIONS Toxic Megacolon and Perforation Toxic megacolon is characterized by gross dilation of the large bowel associated with fever, abdominal pain, dehydration, tachycardia, and bloody diarrhea, which may require urgent surgical intervention. Perforation can occur in the setting of toxic megacolon or in patients with active colitis, especially those taking corticosteroids.

Gastrointestinal Hemorrhage and Anemia Although massive hemorrhage is uncommon, this complication is an indication for surgical intervention. Anemia commonly occurs and is caused by chronic blood loss from the involved colonic mucosa as well as bone marrow suppression from the inflammatory condition.

432

Section VII—Gastrointestinal Disease

Colonic Adenocarcinoma The risk for colon cancer is increased in patients with ulcerative colitis, the magnitude of which is related to the extent and duration of disease. Colon cancer risk is increased 10- to 20-fold (if disease extends proximal to the sigmoid colon) after 8 years of disease compared with that in unaffected individuals. Colonoscopy with surveillance biopsies are recommended every 2 years after 8 to 10 years of disease in patients with pancolitis and after 12 to 15 years in patients with left-sided colitis, followed by yearly examinations after 20 years of disease. Proctitis is not associated with an increased cancer risk. Patients with IBD and primary sclerosing cholangitis (PSC) appear to be at particularly increased risk, and yearly surveillance is recommended after the initial diagnosis of PSC. A minimum of 33 “random” mucosal biopsy samples are recommended during the colonoscopic examination, in addition to targeted samples of circumscript lesions. The use of chromoendoscopy and other enhanced imaging techniques increases the detection of dysplastic lesions in patients with ulcerative colitis. Colectomy is indicated in patients with flat high-grade dysplasia, multifocal flat low-grade dysplasia, possibly unifocal flat low-grade dysplasia, or evidence of colorectal cancer. Polypoid dysplasia entirely removed by polypectomy without flat dysplasia elsewhere in the colon can be managed with continued surveillance colonoscopy.

Clinical Features of Crohn Disease Crohn disease may involve any portion of the gastrointestinal tract, and it is the site of involvement, as well as the type of inflammation, that defines the clinical presentation. Unlike ulcerative colitis, the inflammation in Crohn disease is transmural, and the bowel wall can become thickened, fibrotic, and strictured. The mucosal surface may develop cobblestoning related to edema with linear ulcerations. Deep fissures can develop and result in microperforations and the formation of fistulous tracts. The disease may be continuous but often has skip lesions with intervening segments of normal intestine. The mesentery can become infiltrated with fat, known as creeping fat. The disease is often present for months or years before diagnosis, and, in children, growth retardation may be the sole presenting sign. Distribution of Crohn disease is divided into three major patterns. The most common is ileocecal, which involves the distal portion of the small intestine (terminal ileum) and the proximal large bowel, and is observed in about 40% of patients. Ileocecal Crohn disease may mimic many other diseases, including acute appendicitis. Common symptoms include right lower quadrant abdominal pain, fever, weight loss, and sometimes a palpable inflammatory mass. Chronic inflammation, which leads to fibrosis and stricture formation, may result in partial or complete intestinal obstruction, as demonstrated by abdominal pain, distention, nausea, and vomiting. Because vitamin B12 and bile salts are absorbed in the terminal ileum, ileal Crohn disease or surgical resection of the terminal ileum may lead to B12 deficiency as well as deficiencies of the fat-soluble vitamins (A, D, E, and K) as a result of bile salt malabsorption.

The second major site of Crohn disease involves the small intestine, especially the terminal ileum, and is seen in about 30% of individuals at the time of presentation. Similar complications develop, including fistulas, which may form between different segments of bowel (e.g., enteroenteric, enterocolonic), bowel and skin (enterocutaneous), bowel and bladder (enterovesicular), and bowel and vagina (rectovaginal). The third site of disease is confined to the colon and is observed in 25% of individuals at the time of presentation. Although the disease often spares the rectum, 30% to 40% of patients may develop disabling perianal involvement with fissures, fistulas, and abscesses. Diarrhea is the major consequence but usually with less bleeding than that seen in ulcerative colitis. Distinguishing Crohn colitis from ulcerative colitis can be difficult. The remaining sites of Crohn disease are rare (5%) and include the esophagus, stomach, and duodenum.

MAJOR COMPLICATIONS Stenosis (Stricture) of the Small Intestine or Colon Stenosis may lead to bowel obstruction or stasis with subsequent small intestinal bacterial overgrowth.

Malabsorption Extensive ileal mucosal disease may lead to malabsorption of vitamin B12 (resulting in a megaloblastic anemia and neurologic side effects if not corrected) and malabsorption of bile salts (resulting in diarrhea induced by unabsorbed bile salts and potential fat-soluble vitamin deficiency). Depletion of the bile salt pool can lead to the formation of gallstones. Weight loss may result from generalized malabsorption caused by loss of absorptive surfaces.

Fistulas Transmural inflammation may lead to spontaneous drainage into adjacent bowel loops (enteroenteric fistula), bladder (enterovesical fistula), skin (enterocutaneous fistula), and vagina (rectovaginal), or it may lead to abscess formation around bowel or in other surrounding tissues.

Nephrolithiasis Chronic fat malabsorption leads to luminal binding of free fatty acids to calcium, allowing oxalate, which normally is poorly absorbed because it complexes to calcium in the gut lumen, to be absorbed. This increase in oxalate absorption increases the risk for urinary calcium oxalate stone formation. Patients with an ileostomy or chronic volume loss from diarrhea are also at increased risk for uric acid stones.

Malignancy For colonic Crohn disease, the risk for colorectal cancer is equivalent to ulcerative colitis of similar extent and duration. Therefore, screening and surveillance recommendations are similar to those for ulcerative colitis (see previous discussion). Patients with only small intestinal Crohn disease without colonic involvement are not thought to be at increased risk for colorectal cancer. The rates of small bowel carcinoma and lymphoma are increased in patients with Crohn disease.

Chapter 38—Inflammatory Bowel Disease

Diagnosis The diagnosis of IBD is based on a constellation of clinical features, laboratory tests, and endoscopic, radiographic, and histologic findings. Laboratory tests are not specific and usually reflect inflammation (leukocytosis) or anemia. Perinuclear antineutrophil cytoplasmic antibody (p-ANCA) is positive in up to 70% of patients with ulcerative colitis but rarely positive in patients with Crohn disease, whereas anti– Saccharomyces cerevisiae antibodies (ASCA) are common in Crohn disease and are rarely found in ulcerative colitis. Additional markers including anti–CBir-1 and anti–Omp-C may improve the sensitivity and specificity of serologic testing. Stool examination for ova and parasite identification and testing for Clostridium difficile toxin and enteric bacterial pathogens should be performed to exclude infections that can mimic IBD. Colonoscopy in patients with ulcerative colitis reveals a granular mucosa, decreased vascular markings, decreased mucosal reflection, and superficial ulcerations (Fig. 38-1). In more severe cases, the mucosa is friable, with deeper ulcerations and exudate. Patients with long-standing disease have pseudopolyps, which represent islands of normal tissue in regions of previous ulceration. On endoscopic examination in Crohn disease (Fig. 38-2), the involved mucosa may show aphthoid ulcerations, deep linear or stellate ulcers, edema, erythema, exudate, and friability with intervening areas of normal mucosa (skip lesions). In Crohn disease, small bowel radiography (i.e., small bowel follow-through) has traditionally been the best study with which to investigate the jejunum and ileum, although video capsule endoscopy has recently become increasingly used in this setting. Using this technology, small ulcerations and strictures that are undetectable on small bowel radiography can be visualized (Web Fig. 38-1), although patients

Figure 38-1 Endoscopic image in ulcerative colitis demonstrating diffuse inflammation characterized by erythema, edema, friability, and hemorrhage.

433

with known or suspected strictures or fistulas should not undergo capsule endoscopy given the risk for capsule retention. On small bowel radiography, involved areas have edema and thickening of the bowel wall that lead to bowel loop separation and can also show ulcerations of the mucosa, fistulas, or strictures. A tight, long stricture in the small bowel is commonly called the string sign (Fig. 38-3). Linear ulcers with segments of edematous or uninvolved mucosa lead to the characteristic pattern referred to as cobblestoning. Computed tomographic (CT) scanning can often identify bowel wall thickening with surrounding inflammation as well as intra-abdominal abscesses and fistulas. The recent development of CT enterography and magnetic resonance enterography represent advances in small bowel imaging technology and will likely become primary imaging studies in patients with known or suspected Crohn disease. Mucosal biopsies in ulcerative colitis reveal crypt architectural distortion, with crypt abscesses and infiltration by plasma cells, neutrophils, lymphocytes, and eosinophils (Fig. 38-4). In Crohn disease, the inflammation is transmural and more commonly focal. Granulomas are found in 25% to 30% of histologic specimens in Crohn disease, but not in ulcerative colitis, and can assist in the diagnosis of Crohn disease in the right clinical setting (Fig. 38-5).

Differential Diagnosis The differential diagnosis of IBD includes infectious colitis, ischemic colitis, radiation enteritis, enterocolitis induced by nonsteroidal anti-inflammatory drugs, diverticulitis, appendicitis, gastrointestinal malignancies, and irritable bowel syndrome. In patients with acute onset of bloody diarrhea, infectious causes that must be excluded include Salmonella enteritidis, Shigella species, Campylobacter jejuni, Escherichia

Figure 38-2 Endoscopic image in Crohn disease demonstrating linear ulcers in areas of otherwise normal mucosa.

434

Section VII—Gastrointestinal Disease

Figure 38-4 Mucosal biopsy demonstrating crypt branching and a crypt abscess characteristic of ulcerative colitis (hematoxylin and eosin stain). (Courtesy of Niall Swan, MD.)

Figure 38-3 Radiograph demonstrating small bowel Crohn disease with skip areas and a string sign.

coli O157, and C. difficile. Among the infectious causes, Yersinia enterocolitica can mimic Crohn disease because the pathogen causes ileitis, mesenteric adenitis, fever, diarrhea, and right lower quadrant abdominal pain. Mycobacterium tuberculosis, strongyloidiasis, and amebiasis must be excluded in high-risk populations because these infections can mimic IBD, and treatment with corticosteroids can lead to disseminated infection and death.

Extraintestinal Manifestations Although both ulcerative colitis and Crohn disease primarily involve the bowel, they are associated with inflammatory manifestations in other organ systems, reflecting the systemic nature of these disorders (Table 38-1). Most of these manifestations occur frequently when the bowel is involved, and, in some cases, they may become more difficult to treat than the bowel disease itself. The most common extraintestinal manifestation is arthritis, of which two major types have been identified. The first is a peripheral, large joint, asymmetrical, seronegative, oligoarticular, nondeforming arthritis (about 20% of patients) that may involve the knees, hips, wrists, elbows, and ankles. This peripheral arthropathy usually parallels the course of the large bowel disease (colitic arthritis) and usually lasts for only

Figure 38-5 Colonic biopsy demonstrating chronic inflammatory infiltrate with a granuloma (hematoxylin and eosin stain; magnification 10×). (Courtesy of Niall Swan, MD.)

a few weeks. A second arthritis is axial in location, and its activity does not mirror that of the bowel disease. It consists of sacroiliitis or ankylosing spondylitis. Ankylosing spondylitis (about 5% to 10% of IBD patients) presents with low back pain and stiffness, usually worse during the night, in the morning, or after inactivity. Sacroiliitis alone (without ankylosing spondylitis) is common in IBD (up to about 80% of patients), but many of these patients are asymptomatic. Liver complications of IBD include both intrahepatic and biliary tract diseases. Intrahepatic diseases include fatty liver, pericholangitis, chronic active hepatitis, and cirrhosis. Pericholangitis, also known as small-duct sclerosing cholangitis, is the most common of these diseases and usually is asymptomatic, identified only by abnormalities in alkaline phosphatase and γ-glutamyl transpeptidase on laboratory tests and histologically by portal tract inflammation and bile ductule degeneration. Small-duct sclerosing cholangitis may progress to cirrhosis.

Chapter 38—Inflammatory Bowel Disease Table 38-1 Extraintestinal Manifestations of Inflammatory Bowel Disease Skin Pyoderma gangrenosum Erythema nodosum Sweet syndrome

Table 38-2 Differentiating Features Ulcerative Colitis

Musculoskeletal Seronegative arthritis Ankylosing spondylitis Sacroiliitis Ocular Uveitis Episcleritis

Only involves colon Rectum almost always involved

Pattern of involvement Diarrhea Severe abdominal pain Perianal disease Fistula Endoscopic findings

Continuous

Any area of the gastrointestinal tract Rectum usually spared Skip lesions

Bloody Rare

Usually nonbloody Frequent

No No Erythematous and friable Superficial ulceration Tubular appearance resulting from loss of haustral folds Mucosa only Crypt abscesses

In 30% of patients Yes Aphthoid and deep ulcers Cobblestoning

Miscellaneous Hypercoagulable state Autoimmune hemolytic anemia Amyloidosis

Biliary tract disease includes an increased incidence of gallstones and primary sclerosing cholangitis (PSC). PSC is a chronic cholestatic liver disease marked by fibrosis of the intrahepatic and extrahepatic bile ducts, occurring in 1% to 4% of patients with ulcerative colitis and less often in Crohn disease. Overall, about 70% of patients with PSC have ulcerative colitis. Fibrosis leads to strictures of the bile ducts, which, in turn, may lead to recurrent cholangitis (fever, right upper quadrant pain, and jaundice) and progression to cirrhosis. In addition, about 10% of patients develop cholangiocarcinoma. Medical or surgical therapy for the IBD does not modify the course of PSC, and most patients will pro gress to cirrhosis and liver failure over 5 to 10 years unless a liver transplantation is performed. The two classic dermatologic manifestations of IBD are pyoderma gangrenosum and erythema nodosum. Pyoderma gangrenosum (about 5% of patients) exhibits as a discrete ulcer with a necrotic base, usually on the legs. The ulcer may spread and become large and deep, destroying soft tissues. Pyoderma parallels the activity of the IBD in 50% of cases. Treatment is usually with systemic or intralesional steroids, or both. Other treatment options include dapsone, cyclosporine, and infliximab. Erythema nodosum (10% of patients, usually with peripheral arthropathy) exhibits raised, tender nodules, usually over the anterior surface of the tibia. It heals without scarring and responds to treatment for the underlying bowel disease. A less common dermatologic manifestation of IBD is Sweet syndrome, or acute febrile neutrophilic dermatosis. This is a condition characterized by the sudden onset of fever, leukocytosis, and tender, erythematous, welldemarcated papules and plaques that show dense neutrophilic infiltrates on histologic examination. Ocular manifestations of IBD include uveitis and episcleritis (5%). Uveitis (or iritis) is an inflammatory lesion of the anterior chamber and produces blurred vision, photophobia, headache, and conjunctival injection. Local therapy includes steroids and atropine. Episcleritis is less serious

Crohn Disease

Site of involvement

Hepatobiliary Primary sclerosing cholangitis Cholelithiasis Autoimmune hepatitis

435

Radiologic findings

Histologic features Smoking Serology

Protective p-ANCA more common

String sign of terminal ileum RLQ mass, fistulas, abscesses Transmural Crypt abscesses, granulomas (about 30%) Worsens course ASCA more common

ASCA, anti–Saccharomyces cerevisiae antibodies; p-ANCA, perinuclear antineutrophil cytoplasmic antibody; RLQ, right lower quadrant.

than uveitis, producing burning eyes and scleral injection, and is treated with topical steroids. Other complications of IBD include chronic anemia (common), digital clubbing and hypertrophic osteoarthropathy (uncommon in adults), an increased incidence of thromboembolic disease (uncommon), and amyloidosis (rare).

Differentiation between Ulcerative Colitis and Crohn Disease Generally, the diagnoses of ulcerative colitis and Crohn disease can be made based on the findings as described in their respective sections presented earlier and those outlined in Table 38-2. As noted, most Crohn patients have small bowel involvement, skip lesions, and pain, whereas most ulcerative colitis patients have bloody diarrhea with involvement of the rectum and a continuous, superficial spread of the disease. The endoscopic, radiologic, and histologic criteria aid in the phenotypic differentiation of these disease entities. However, occasionally, a diagnosis of indeterminate colitis is made as a result of an overlap of findings. For example, colonic Crohn disease may produce superficial continuous rectal involvement similar to ulcerative proctitis. Similarly, chronic ulcerative colitis can infrequently result in

436

Section VII—Gastrointestinal Disease

inflammation of the terminal ileum, called backwash ileitis. In many indeterminate cases, repeated examination is necessary, or complications develop that help identify the form of the disease.

Treatment (Induction and Maintenance of Remission) As part of the initial management of patients with IBD, the clinician must determine the extent and assess the severity of the disease. Patients with mild or moderate disease can be managed as outpatients with close monitoring in association with a gastroenterologist. Patients with severe or fulminant disease, as indicated by abdominal pain, fever, tachycardia, anemia, and leukocytosis, require hospital admission and multidisciplinary team management. Because IBD is a chronic recurrent illness, treatment is centered on controlling the acute attack with induction of remission followed by maintenance of remission. Treatment options for ulcerative colitis and Crohn disease are reviewed in Table 38-3.

5-AMINOSALICYLIC ACID The aminosalicylates are given either orally or topically (suppository and enema) and are safe and effective in the treatment of mild to moderate disease as well as in maintenance of remission of ulcerative colitis. The efficacy of most of these agents in the induction or maintenance of remission of Crohn disease is questionable. This category includes sulfasalazine (Azulfidine) at a dose of 4 to 6 g/day in divided doses, which consists of 5-aminosalicylic acid (5-ASA) linked to a sulfapyridine moiety and which is activated following the release of the 5-ASA after bacterial lysis in the colon. Side effects, including headache, nausea, and skin

Table 38-3 Treatment Options Disease Severity

Ulcerative Colitis

Mild

Oral and topical 5-ASA compounds Oral and topical 5-ASA compounds Oral steroids Azathioprine, 6-MP Infliximab

Moderate

Severe

Intravenous steroids Cyclosporine Infliximab Surgery

Crohn Disease 5-ASA compounds Antibiotics Elemental diet 5-ASA compounds Antibiotics Budesonide or oral steroids Azathioprine, 6-MP Methotrexate Infliximab, adalimumab, certolizumab, natalizumab Intravenous steroids Azathioprine, 6-MP Methotrexate Infliximab, adalimumab, certolizumab, natalizumab Surgery

5-ASA, 5-aminosalicylic acid; 6-MP, 6-mercaptopurine.

reactions, may require discontinuation of sulfasalazine in about 30% of patients. Reversible oligospermia may occur with sulfasalazine, and rare serious side effects include pleuropericarditis, pancreatitis, agranulocytosis, interstitial nephritis, and hemolytic anemia. Patients who take sulfasalazine need folic acid supplementation. Newer derivatives of oral 5-ASA compounds, such as mesalamine (Pentasa, 4g per day in divided doses; Asacol, 2.4g per day in divided doses; Lialda, 2.4 to 4.8g once daily), olsalazine (Dipentum, 1 to 2g per day in divided doses), and balsalazide (Colazal, 6.75 g per day in divided doses), as well as topical forms of mesalamine (Canasa suppositories, 1000mg once daily; or Rowasa enemas, 4g once nightly), are being commonly used because of a favorable side-effect profile. In addition to their use in the primary treatment of IBD, several studies suggest that long-term use of 5-ASA medications may reduce the risk for colorectal cancer in patients with ulcerative colitis.

CORTICOSTEROIDS Corticosteroids may be used topically, orally, or intravenously and are effective for controlling active disease but are not useful for maintaining remission. They are indicated for moderate or severe disease and in patients in whom treatment with 5-ASA fails. The most commonly used agent is prednisone, started in doses between 40 and 60mg per day. Patients typically improve rapidly, and the medication is usually tapered down slowly, that is, 5 to 10 mg per week until discontinuation. Patients who do not improve after 1 week of oral treatment and those with more severe disease are best treated in the hospital with intravenous corticosteroids, such as intravenous hydrocortisone, 300mg per day, or methylprednisolone, which can be given either by continuous infusion or in three divided doses. Corticosteroids have numerous side effects with long-term use. Budesonide (Entocort, 9 mg given once daily), a corticosteroid that undergoes extensive first-pass hepatic metabolism, is now available for inducing and maintaining remission of ileocolonic Crohn disease and may offer long-term benefits with decreased corticosteroid side effects. Controlled trials have shown that budesonide is more effective than placebo and oral 5-ASA, and has similar efficacy to prednisolone for the induction of remission in Crohn disease.

ANTIBIOTICS Antibiotics are primarily used in patients with Crohn disease who have colonic, perianal, or fistulizing disease. Intravenous antibiotics are also part of the initial treatment in patients with severe, toxic, or fulminant colitis. The two commonly used antibiotics are metronidazole (Flagyl) and ciprofloxacin (Cipro). Ciprofloxacin is prescribed at a dosage of 500 mg twice a day. Metronidazole is prescribed at a dosage of 20mg/kg per day in three divided doses. Patients should be warned of potential side effects, such as a disulfiram (Antabuse) effect and peripheral neuropathy.

IMMUNOMODULATORS Included in this category are azathioprine (Imuran, 2 to 2.5 mg/kg per day) and its active metabolite 6-

Chapter 38—Inflammatory Bowel Disease mercaptopurine (6-MP) (Purinethol, 1 to 1.5 mg/kg per day) as well as methotrexate and cyclosporine. Azathioprine and 6-MP are effective therapies for maintaining remission in both Crohn disease and ulcerative colitis and are used primarily as steroid-sparing agents. They have a slow onset of action (months) but are generally safe and well tolerated. Other regimens include subcutaneous or intramuscular methotrexate for induction (25 mg weekly) and maintenance of remission (15 to 25 mg weekly) in active Crohn disease and intravenous cyclosporine (2 to 4mg/kg per day given over 24 hours) as bridge treatment for severe steroidrefractory ulcerative colitis. Given the potential for both short-term and long-term side effects, as well as the need for close follow-up, patients needing these medications are best managed by gastroenterologists.

BIOLOGIC THERAPY Advances in our knowledge of the immunopathogenesis of IBD have led to a new class of therapies that target specific aspects of the immune system, collectively known as the biologic agents. The first such agent to be used in IBD was infliximab, a chimeric monoclonal antibody to TNF-α, which has been shown to be effective in the treatment of moderate to severe Crohn disease, including fistulizing disease. More recently, infliximab has also been shown to be beneficial in the treatment of ulcerative colitis that is refractory to conventional medical therapy. Because infliximab is a chimeric antibody, its toxicities include infusion reactions, delayed-type hypersensitivity reactions, and formation of autoantibodies (which can reduce its efficacy). Two newer anti-TNF agents, adalimumab (a fully human monoclonal antibody), and certolizumab pegol (a humanized anti-TNF antibody Fab fragment), have also proved efficacious in patients with moderate to severe Crohn disease who have not responded well to conventional treatments. These two agents are administered subcutaneously, and may have a lower toxicity profile. Natalizumab, a humanized anti–α4integrin antibody, blocks inflammatory cell migration and adhesion and has recently been approved for the treatment of moderate to severe Crohn disease in patients who have had an inadequate response to, or are unable to tolerate, conventional Crohn disease therapies including inhibitors of TNF-α. Because of the potent effects these biologic agents have on the immune system, careful patient selection and monitoring for complications are necessary. Reactivation of latent tuberculosis and other serious infections have been reported with the anti-TNF agents, and there may be an increased risk for non-Hodgkin lymphoma and possibly solid tumors as well. Natalizumab has been linked to rare cases of progressive multifocal leukoencephalopathy caused by the human polyoma JC virus.

PROBIOTICS Probiotics are viable nonpathogenic organisms that, after ingestion, may prevent or treat intestinal diseases. Probiotics are being explored in treatment of IBD and may help prevent recurrence after surgery for Crohn disease and to treat pouchitis after ileal pouch-anal anastomosis.

437

NUTRITIONAL SUPPORT Nutritional support is an important adjunctive aspect in the management of patients with IBD. However, the role of nutrition as a primary treatment has been limited to patients with small bowel Crohn disease. These patients may achieve and maintain remission with total parenteral nutrition or elemental diets after prolonged periods (at least 4 weeks). Many patients with Crohn disease and ulcerative colitis experience weight loss during exacerbations of their illness and need caloric supplements. Vitamins and minerals can be given orally as a multivitamin with folic acid. Vitamin B12 should be supplemented parenterally in patients who have extensive ileal disease or an ileal resection. Patients taking corticosteroids require supplemental calcium and vitamin D, and individuals with extensive small bowel involvement can also develop malabsorption of fat-soluble vitamins (A, D, E, and K), iron deficiency, and rarely trace minerals. Lactose-free diets, as well as low-fiber diets, may be necessary in patients with active disease or strictures.

ANTIDIARRHEALS AND BILE SALT RESIN BINDERS Antidiarrheal agents and bile salt resin binders are adjuncts used to manage diarrhea in patients with IBD. Antidiarrheal agents should be used cautiously during exacerbations of colitis because they can precipitate toxic megacolon. The main role of these medications involves controlling diarrhea in patients who have undergone previous resections. Generally, when less than 100 cm of terminal ileum has been resected, patients can develop a bile salt malabsorptive state during which bile salts enter the colon and result in a secretory diarrhea. Bile salt resin binders such as cholestyramine are an effective treatment in these cases. When patients have undergone one or more extensive resections, the bile salt pool is depleted, and fat malabsorption develops. These patients may require a low-fat diet supplemented with medium-chain triglycerides and antidiarrheal agents, but bile salt resin binders should not be used.

SURGICAL MANAGEMENT Surgical intervention is indicated for patients with severe complications such as obstruction, perforation, massive gastrointestinal hemorrhage, and toxic megacolon not responsive to medical treatment. The other main indication for surgical treatment is the presence of dysplasia or cancer. For patients with ulcerative colitis, regardless of the extent of disease, the entire colon must be removed, and the operation is essentially curative. About 20% to 25% of patients have pancolitis, and one third to one half will require colectomy within 2 to 5 years of diagnosis, depending on the severity of their colitis. In contrast, less than 10% of individuals with mild disease or proctitis will undergo colectomy by 10 years after diagnosis. Historically, the initial operation for ulcerative colitis was a total proctocolectomy and Brooke ileostomy. More recently, the ileal pouch-anal anastomosis has become the operation of choice in most patients. In this operation, the colon is removed, and the small bowel is constructed into a reservoir (ileal pouch) that

438

Section VII—Gastrointestinal Disease

is anastomosed to the anus, allowing defecation through the anus. Complications of this operation include the development of pouchitis, fecal incontinence, reduced fertility, and need for reoperation. Surgery is not curative in Crohn disease and is generally avoided, if possible. Nonetheless, 10 years after a diagnosis of Crohn disease, more than 60% of patients will require

surgery. Many surgical procedures in patients with Crohn disease are performed to manage complications of the disease, including segmental resection, stricturoplasty, fistulectomy, and abscess drainage. Unfortunately, the recurrence rate is high, with 70% of patients having an endoscopic recurrence within 1 year of surgery and 50% having a symptomatic recurrence within 4 years.

Prospectus for the Future As our understanding of the etiologic and pathophysiologic aspects of inflammatory bowel disease increases, major advancements in diagnosis and treatment are anticipated. These advancements include the following: • The use of molecular, genetic, and serologic tests to differentiate between subtypes of disease as well as identify individuals at high risk for developing complications of inflammatory bowel disease

References Baumgart DC, Sandborn WJ: Inflammatory bowel disease: Clinical aspects and established and evolving therapies. Lancet 369:1641, 2007. Brown SJ, Mayer L: The immune response in inflammatory bowel disease. Am J Gastroenterol 102:2058, 2007. Cho JH: The genetics and immunopathogenesis of inflammatory bowel disease. Nat Rev Immunol 8:458, 2008. Cima RR, Pemberton JH: Medical and surgical management of chronic ulcerative colitis. Arch Surg 140:300, 2005. D’Haens G, Baert F, van Assche G, et al: Early combined immunosuppression or conventional management in patients with newly diagnosed Crohn’s disease: An open randomised trial. Lancet 371:660-667, 2008. Itzkowitz SH, Present DH: Crohn’s and Colitis Foundation of America, Colon Cancer in IBD Study Group: Consensus Conference. Colorectal cancer

• The increased, and earlier, use of biologic agents to specifically target aspects of the immune system and inflammatory pathways known to be involved in IBD pathophysiology • Improvements in the detection of dysplasia and prevention of colorectal cancer (including chemoprevention) in patients with chronic colitis

screening and surveillance in inflammatory bowel disease. Inflamm Bowel Dis 11:314, 2005. Katz S: Update in medical therapy of ulcerative colitis: Newer concepts and therapies. J Clin Gastroenterol 39:557, 2005. Kornbluth A, Sachar DB: Ulcerative colitis practice guidelines in adults (update): American College of Gastroenterology, Practice Parameters Committee. Am J Gastroenterol 99:1371, 2004. Peyrin-Biroulet L, Desreumaux P, Sandborn WJ, Colombel JF: Crohn’s disease: Beyond antagonists of tumour necrosis factor. Lancet 372:67-81, 2008. Sandborn WJ, Feagan BG, Lichtenstein GR: Medical management of mild to moderate Crohn’s disease: Evidence-based treatment algorithms for induction and maintenance of remission. Aliment Pharmacol Ther 26:987-1003, 2007. Siegel CA, Sands BE: Review article: Practical management of inflammatory bowel disease patients taking immunomodulators. Aliment Pharmacol Ther 22:1, 2005.

Chapter

VII

Neoplasms of the Gastrointestinal Tract Paul C. Schroy III

Esophageal Carcinoma Carcinoma of the esophagus is one of the most lethal of all cancers. The lack of early symptoms and serosal barrier, as well as the rich, bidirectional esophageal lymphatic flow, often results in advanced disease by the time of diagnosis. The American Cancer Society estimates that about 16,500 new cases of esophageal cancer and 14,530 esophageal cancer deaths will occur in the United States in 2009. Historically, squamous cell carcinoma (SCC) constituted 95% of all esophageal carcinomas. Since 1980, however, the incidence of adenocarcinoma of the esophagus has rapidly increased and now accounts for about 50% of newly diagnosed cases of esophageal carcinoma. The epidemiology of SCC differs from that of adenocarcinoma of the esophagus, but the symptoms, treatments, and prognoses are similar.

INCIDENCE AND EPIDEMIOLOGY The incidence of SCC varies dramatically throughout the world. The highest rates are found in developing countries such as northern China, Iran, India, and parts of southern Africa. SCC is relatively uncommon in the United States, with an annual incidence of less than 5 cases per 100,000 population. Esophageal cancer is rare among individuals younger than 40 years, but thereafter increases in incidence with each subsequent decade. Men are affected more often than women, and African Americans have a fivefold increase in incidence compared with other racial and ethnic groups. The cause of SCC is unknown, but environmental, dietary, and local esophageal factors have been implicated. Heavy alcohol consumption and smoking are the predominant risk factors for SCC in the United States. In developing countries, nutritional deficiencies (e.g., selenium), betel nut chewing, human papillomavirus infection, and consumption of extremely hot drinks (e.g., tea), nitrates, and pickled vegetables, are also important risk factors. Predisposing conditions include lye strictures, radiation injury, Plummer-Vinson syndrome, achalasia, tylosis, and celiac disease.

Adenocarcinoma of the esophagus is primarily a disease of white men. The primary risk factor for adenocarcinoma is Barrett esophagus, a condition in which specialized intestinal-type columnar mucosa replaces the normal squamous mucosa in response to chronic gastroesophageal reflux disease. It is presumed that intestinal metaplasia progresses to low-grade dysplasia and then high-grade dysplasia and finally adenocarcinoma. The risk for developing adenocarcinoma in the setting of Barrett esophagus is about 0.5% per year. Long-standing gastroesophageal disease, obesity, and cigarette smoking have also been implicated as potential causative factors. Endoscopic surveillance with biopsy is recommended for Barrett esophagus but not chronic gastroesophageal reflux disease.

Clinical Presentation Early and curable esophageal carcinoma is frequently asymptomatic and detected serendipitously. The presence of symptoms heralds an advanced and most often incurable stage of disease. Under careful questioning, most patients will have had symptoms for a few months before seeking medical attention. Dysphagia is the most common symptom of esophageal carcinoma. It occurs when the esophageal lumen has been compromised by about 75% of its normal diameter. Difficulty swallowing solid foods precedes dysphagia to liquids. With complete obstruction, regurgitation, aspiration, and cough or pneumonia may occur. Pulmonary symptoms may also occur if a tracheoesophageal fistula is present. Patients uniformly have weight loss and anorexia. Chest pain, hiccups, or hoarseness indicates involvement of adjacent structures such as the mediastinum, diaphragm, and recurrent laryngeal nerve, respectively. If gastrointestinal bleeding occurs, it is often occult or associated with iron deficiency anemia. Life-threatening gastrointestinal hemorrhage can occur if the tumor has invaded major vessels. Clubbing of the nails and paraneoplastic syndromes, such as hypercalcemia and Cushing syndrome, are rarely seen. 439

440

Section VII—Gastrointestinal Disease H. pylori

DIAGNOSIS Patients with dysphagia or other suggestive symptoms should be evaluated by upper endoscopy or an esophageal barium study. The advantage of endoscopy includes the opportunity to obtain tissue of the cancer, either by biopsy or brush cytologic study. Esophageal carcinoma may appear as a plaque, an ulcer, a stricture, or a mass. Nearly 90% of adenocarcinomas develop in the distal esophagus, whereas 50% of SCCs occur in the middle third of the esophagus, and the other 50% are evenly distributed in the proximal and distal esophagus. Computed tomography (CT) scanning of the chest and abdomen is performed to detect invasion of local structures and metastases to the lung and liver. Endoscopic ultrasonography (EUS), with its ability to image the esophageal wall as a five-layer structure that correlates with histologic layers, is more accurate than CT for staging tumor depth, local invasion, and regional node involvement. EUS also permits targeted fine-needle aspiration of suspicious findings.

THERAPY Stage is the most important prognostic factor for the survival of patients with esophageal cancer and influences the treatment options. Staging is based on the tumor-node-metastasis (TNM) classification system. Only localized tumors confined to the wall of the esophagus are potentially curable by surgery. Overall 5-year survival rates for patients undergoing curative resection, however, are just 5% to 20%. Preoperative chemotherapy with multidrug regimens combined with radiation therapy may reduce local recurrence rates and improve survival. Chemotherapy plus radiation therapy is also recommended for patients with locally unresectable disease, medical conditions that preclude surgery, and those who refuse surgery. Patients with metastatic disease should be considered for palliative treatment of dysphagia. Local treatment with endoscopic methods (such as malignant stricture dilation), placement of an endoprosthesis (stent), and tumor ablation by laser or photodynamic therapy are often the methods of choice for rapid palliation. More sustained palliation can be achieved using combined chemotherapy and radiation therapy.

Gastric Carcinoma Gastric carcinoma is one of the leading causes of cancerrelated deaths worldwide. For unknown reasons, the incidence of gastric cancer has declined dramatically in the United States since the 1930s. Despite its declining incidence, the American Cancer Society estimates that about 21,100 new cases and 10,620 gastric cancer deaths will occur in 2009. Unfortunately, gastric cancer is often advanced at the time of diagnosis; the 5-year survival rate is about 24%.

INCIDENCE AND EPIDEMIOLOGY More than 90% of gastric cancers are adenocarcinomas. The incidence of gastric cancer varies widely throughout the world. The disease is more common in developing countries than industrialized nations and shows a predilection for

Dietary factors

Chronic active gastritis

Host factors

Intestinal metaplasia

Dysplasia

ADENOCARCINOMA Figure 39-1 Model for the development of gastric adenocarcinoma.

urban and lower socioeconomic groups. Japan, China, the Andean regions of South America, and Eastern Europe exhibit the highest rates. The United States has among the lowest incidence rates at less than 10 cases per 100,000 population. Gastric cancer rarely occurs before age 40 years; thereafter, the incidence rises steadily, peaking in the seventh decade. Men are afflicted at a rate nearly twice that of women. African Americans, Hispanic Americans, and Native Americans are 1.5 to 2.5 times more likely to develop gastric cancer than whites. Migrants typically acquire the risk of their host countries, suggesting an important role for environmental factors. Low socioeconomic status, improper food storage, and other dietary and local gastric factors are associated with the disease. Dietary factors include deficiencies in fats, protein, and vitamins A and C and excesses in salted meat and fish, smoked foods, pickled vegetables, and nitrates. Predisposing conditions including atrophic gastritis, postgastrectomy states, achlorhydria, pernicious anemia, adenomatous polyps, and Ménétrier disease are also associated with an increased incidence. The World Health Organization has classified Helicobacter pylori as a carcinogen and epidemiologically linked to gastric adenocarcinoma (Fig. 39-1). However, only a small proportion of patients infected with H. pylori develop gastric adenocarcinoma. Gastric lymphomas account for fewer than 5% of primary gastric malignancies. The stomach is the most common site of extranodal non-Hodgkin lymphoma, but Hodgkin lymphoma of the stomach is rare. Gastric mucosa-associated lymphoid tissue (MALT) lymphomas are associated with H. pylori infection in 90% of cases and are reported to regress in 60% to 70% of cases after eradication of H. pylori. MALT lymphomas can also occur in association with various autoimmune and immunodeficiency syndromes. Most develop in individuals older than 50 years, and there is a slight male predominance.

CLINICAL PRESENTATION The location, size, and growth pattern of gastric malignancies may influence the presenting symptoms. Abdominal discomfort is the most frequent symptom; however, early satiety, nausea, and vomiting may occur, especially with gastric outlet obstruction. Gastrointestinal bleeding may manifest as iron deficiency anemia, occult bleeding, or

Chapter 39—Neoplasms of the Gastrointestinal Tract frank upper gastrointestinal hemorrhage. Anorexia and weight loss often accompany other symptoms. The signs of metastatic disease, which may be found on physical examination and signify incurability, include a Virchow (left supraclavicular) node, a Blumer shelf (mass in the perirectal pouch, found on digital rectal examination), and a Krukenberg tumor (metastasis to the ovaries). A variety of paraneoplastic syndromes have been associated with gastric adenocarcinoma and warrant an investigation for a gastrointestinal malignancy. They include Trousseau syndrome (thrombosis), acanthosis nigricans (pigmented dermal lesions), membranous nephropathy, microangiopathic hemolytic anemia, Leser-Trélat sign (seborrheic keratoses), and dermatomyositis.

DIAGNOSIS The diagnostic tests for gastric malignancies include double contrast (barium) upper gastrointestinal radiography or endoscopy. Lesions detected on barium study require endoscopic biopsy and cytologic study for histologic evaluation. Gastric carcinomas may appear as ulcers, masses, enlarged gastric folds, or an infiltrative process with a nondistensible stomach wall (linitis plastica). The accuracy of endoscopic ultrasonography is in the range of 77% to 93% for determining the depth of invasion and 65% to 90% for predicting regional node involvement. CT scanning of the chest and abdomen may detect metastases in the lung and liver but is otherwise poor for staging. Laparoscopy is increasingly being used for staging and determination of resectability with high accuracy.

THERAPY The standard treatment of gastric cancer is complete surgical resection with removal of all gross and microscopic disease. The postoperative local-regional recurrence rate remains 80%. A postoperative combination of chemotherapy plus radiation therapy reduces local recurrence rates and improves survival in patients undergoing curative resection. In the United States, nearly two thirds of patients present with advanced disease (stages III to IV), with a survival rate of less than 20%. Chemotherapy is the mainstay of treatment for such patients, but long-term survival is rare. Palliative resection may be performed to prevent obstruction or treat bleeding; radiation therapy and endoscopy may also be of palliative benefit in select patients. Treatment options for gastric lymphomas include some combination of chemotherapy, radiation therapy, and surgery, depending on stage of disease.

Colorectal Polyps and Carcinoma Carcinoma of the colon and rectum is the third most common cancer and the second most common cause of cancer deaths in American men and women. More than 146,970 new cases and 49,920 colorectal cancer-related deaths will occur in 2009. Screening has been shown to be an effective strategy for reducing both colorectal cancer mortality, through early detection, and incidence, through the identification and removal of premalignant adenomas.

441

INCIDENCE AND EPIDEMIOLOGY Worldwide incidence and mortality of colorectal cancer varies considerably. With the notable exception of Japan, industrialized countries are at greatest risk. In the United States, incidence rates have declined slightly during the past decade but remain in excess of 40 cases per 100,000 persons. About 6% of Americans will develop colorectal cancer during their lifetime. Age is an important determinant of risk. Although extremely uncommon in individuals younger than 35 years (except those with rare predisposing genetic syndromes), the incidence of colorectal cancer increases steadily with age, beginning at about 40 years of age, with an approximate doubling with each successive decade thereafter to about 80 years of age. Cancer of the colon affects men and women at similar rates, whereas cancer of the rectum is more common in men. Colorectal cancer does not appear to have a racial predilection; however, African Americans are more likely to present with advanced-stage disease. Epidemiologic studies have identified a number of modifiable risk factors related to colorectal cancer. Factors associated with an increased risk for the disease include obesity, red meat, alcohol, and tobacco; conversely, factors associated with a decreased risk include physical activity, nonsteroidal antiinflammatory agents, and multivitamins. Most colorectal cancers are believed to arise from benign adenomatous polyps (adenomas). The epidemiology of colorectal adenomas is similar to that of colorectal cancer. In general, the prevalence of colorectal adenomas in a given country parallels the prevalence of colorectal cancer. Age is an important determinant of prevalence in high-risk countries. In the United States, autopsy studies suggest an overall prevalence of 50%, ranging from about 30% at age 50 years to 55% at age 80 years. Fortunately, only a minority of adenomas progress to colorectal cancer. It is unknown how long an adenoma takes to develop into an invasive cancer, but data from multiple observational studies suggest at least 10 years. Insight into the molecular mechanisms responsible for the adenoma-carcinoma sequence suggests that colorectal carcinogenesis is a multistage process (Fig. 39-2) resulting from the accumulation of genetic alterations involving various oncogenes (e.g., K-ras), tumor suppressor genes (e.g., APC or β-catenin, DCC, SMAD-4, SMAD-2, and p53), or DNA mismatch repair genes (e.g., hMLH-1). High-risk groups have been identified and include those with a personal or family history of colorectal cancer or adenomas, various genetic polyposis and nonpolyposis syndromes, and inflammatory bowel disease (Table 39-1). Hereditary nonpolyposis colorectal cancer (HNPCC) and familial adenomatous polyposis (FAP) are well-defined genetic syndromes associated with the highest risk for colorectal cancer. HNPCC (Lynch syndromes) is characterized by inherited mutations in one of the DNA mismatch repair genes (e.g., hMLH-1 or hMSH-2), early-onset colorectal cancer (average age, 44 years) in the absence of polyposis, a predominance (60% to 70%) of tumors proximal to the splenic flexure, an excess of both colorectal and extracolonic (e.g., endometrial) cancers, and an estimated lifetime risk for colorectal cancer of 80% to 90%. In contrast, FAP is characterized by inherited mutations in the APC gene, the appearance of hundreds of colorectal adenomas during the second or third decade of life, and a risk for colorectal cancer

442

Section VII—Gastrointestinal Disease DNA hypomethylation

Chromosome alteration gene

12p mutation K-ras

5q mutation or loss APC/b-catenin

Normal epithelium

Hyperproliferative epithelium

Other alterations

Early adenoma

18q loss DCC/SMAD-2/SMAD-4

Intermediate adenoma

Late adenoma

17p loss p53

Carcinoma

Metastasis

DNA Mismatch Repair (MMR) gene inactivation (e.g., MLH-1 hypermethylation) Figure 39-2 A genetic model for colorectal tumorigenesis.

Table 39-1 Risk Factors for Colorectal Cancer Age ≥ 50yr Personal history of adenomatous polyps or colorectal cancer Familial adenomatous polyposis or Gardner syndrome MYH-associated adenomatous polyposis Hereditary nonpolyposis colon cancer Ulcerative colitis or Crohn colitis First-degree relative with colon cancer or adenomatous polyps diagnosed before age 60yr Hamartomatous polyposis syndrome (Peutz-Jeghers syndrome, juvenile polyposis)

that approaches 100% by the fifth decade if left untreated. FAP is also associated with benign fundic gland polyps in the stomach and duodenal adenomas and adenocarcinomas that have a predilection for the periampullary region. Gardner syndrome is a variant of FAP in which affected probands also exhibit a variety of extraintestinal manifestations such as osteomas, desmoids, and other soft tissue tumors. Congenital hypertrophy of the retinal pigment epithelium is an early benign manifestation of both FAP and Gardner syndrome. MYH-associated adenomatous polyposis syndrome is indistinguishable from FAP clinically but caused by mutations of the base excision repair gene, mutY homologue (MYH) rather than APC. Peutz-Jeghers syndrome is an autosomal dominant condition characterized by hamartomatous polyposis of both the small and large bowel and mucocutaneous pigmentation. Affected individuals are at increased risk for both gastrointestinal (stomach, small bowel, and colon) and extraintestinal (e.g., genital tract, pancreas, and breast) malignancies occurring at a young age. Generalized juvenile polyposis is another inherited hamartomatous polyposis syndrome associated with a small, albeit increased, risk for colorectal cancer.

CLINICAL PRESENTATION Most colorectal neoplasms are asymptomatic until advanced. Gastrointestinal blood loss is the most common symptom

and may present as occult bleeding, hematochezia, or unexplained iron deficiency anemia. Other symptoms include abdominal pain from obstruction or invasion, change in bowel habits, or unexplained anorexia or weight loss. A palpable mass may be present in patients with advanced cancers of the cecum.

DIAGNOSIS All patients with symptoms suggestive of colorectal neoplasia should undergo an evaluation of the colon by colonoscopy, flexible sigmoidoscopy, or double contrast barium enema. About 50% of colorectal adenomas and cancers are located between the rectum and splenic flexure; however, the prevalence of cancers proximal to the splenic flexure increases with increasing age, especially among women. Colorectal cancers may arise in sessile (flat) or pedunculated (on a stalk) polyps, or they may appear as a stricture, a fungating mass, or an ulcerated mass. Colonoscopy has greater accuracy than a barium enema study in the detection of small polyps and early cancers as well as the ability to remove neoplasms or biopsy lesions at the time of the examination. Lesions detected on barium enema study necessitate colonoscopic evaluation. CT scanning and of the abdomen and pelvis is used preoperatively to assess the extent of metastatic disease. Magnetic resonance scanning and positron emission tomography may also be useful in detecting metastatic disease in select patients. EUS is used for the preoperative staging of rectal cancer. Carcinoembryonic antigen level is measured preoperatively for a baseline value and, if elevated, monitored to detect tumor recurrence postoperatively. Periodic screening by colonoscopy, CT colonography (virtual colonoscopy), flexible sigmoidoscopy, or doublecontrast enema is recommended for asymptomatic, averagerisk patients beginning at age 50 years. Stool blood testing and stool DNA testing are alternative screening methods for patients who refuse one of the preferred methods (Table 39-2). Screening recommendations for high-risk patients vary depending on the risk factor (see Table 39-2) but in general rely on colonoscopy performed at a younger age and at more frequent intervals than for those at average risk.

Chapter 39—Neoplasms of the Gastrointestinal Tract

443

Table 39-2 Colorectal Cancer (CRC) Screening and Surveillance Recommendations* Indication

Recommendations

Average risk

Beginning at age 50yr: Colonoscopy every 10yr Computed tomographic colonography every 5yr Flexible sigmoidoscopy every 5yr Double-contrast barium enema every 5yr (Stool blood testing annually or stool DNA testing acceptable but not preferred) Colonoscopy every 5yr beginning at age 40yr, or 10yr younger than earliest diagnosis, whichever comes first Genetic counseling and screening† Colonoscopy every 1 to 2 years beginning at age 25yr and then yearly after age 40yr‡ Genetic counseling and testing† Flexible sigmoidoscopy yearly beginning at puberty‡ Colonoscopy within 1yr of curative resection; repeat at 3yr and then every 5yr if normal Colonoscopy every 3 to 5yr after removal of all index polyps Colonoscopy every 1 to 2yr beginning after 8yr of pancolitis or after 15yr if only left-sided disease

One or two first-degree relatives with CRC at any age or adenoma at age < 60yr Hereditary nonpolyposis colorectal cancer

Familial adenomatous polyposis and variants Personal history of CRC Personal history of colorectal adenoma Inflammatory bowel disease

*Recommendations proposed by the American Cancer Society and U.S. Multi-Society Task Force on Colorectal Cancer; recommendations for averagerisk patients also endorsed by the American College of Radiology. † Whenever possible, affected relatives should be tested first because of potential false-negative results. ‡ Screening recommendation for individuals with positive or indeterminate tests as well as for those who refuse genetic testing.

Colonoscopic surveillance is recommended for patients with a history of colorectal cancer or adenomas and inflammatory bowel disease.

THERAPY The rate of survival of patients with colorectal carcinoma is based on the stage of disease (Table 39-3). Unfortunately, 45% of patients first come to medical attention with stage III or IV disease. Surgery alone is curative for early-stage colorectal cancers. Surgery and adjuvant chemotherapy with 5-fluorouracil and leucovorin ± oxiliplatin or capecitabine alone are recommended for stage III colon cancer. For patients with stage II and III rectal cancer, the combination of postoperative radiation and 5-fluorouracil (± leucovorin) has been found to significantly reduce the recurrence rate, cancer-related deaths, and overall mortality. Independent of nodal status, preoperative chemoradiotherapy followed adjuvant chemotherapy is recommended for patients with locally advanced rectal cancers. For patients with stage IV disease, palliative surgery, chemotherapy, and radiation therapy are the mainstays of therapy.

Carcinoid Tumors The overall incidence of gastrointestinal carcinoid tumors in the United States is estimated at 1 to 2 cases per 100,000 people. The most common sites, in descending order of frequency, are the small intestine (ileum), rectum, appendix, colon, and stomach. Carcinoid tumors arise from neuroendocrine cells and contain a variety of secretory granules containing various hormones and biogenic amines. Serotonin is synthesized

Table 39-3 Survival and Comparison of Dukes and TNM Staging in Colorectal Carcinoma Dukes A B C D

TNM Stage

5-Year Survival Rate (%)

I II III IV

93 72-85 44-83 8

from 5-hydroxytryptophan and metabolized in the liver to 5-hydroxyindoleacetic acid, which is biologically inert and secreted in the urine. The release of serotonin (hindgut tumors) and other vasoactive substances into the systemic circulation is thought to cause the carcinoid syndrome. Therefore, carcinoid metastases in the liver or other sites that drain into systemic veins may be associated with the carcinoid syndrome, as may primary carcinoids in the ovary or bronchus. The symptoms include episodic flushing, wheezing, diarrhea, right-sided valvular heart disease, and, potentially, vasomotor collapse. Localized tumors may present with gross or occult bleeding, obstructive symptoms, or abdominal pain depending on their location. Most carcinoids are indolent; however, the malignant potential is variable and appears to be related to the site and, often, the size of the primary tumor. Carcinoids arising in the ileum and those 2cm or larger have the greatest malignant potential. Surgical resection is the only curative treatment for carcinoid tumors. Somatostatin analogues are highly effective in the management of the symptoms of carcinoid syndrome.

444

Section VII—Gastrointestinal Disease

Prospectus for the Future Further elucidation of the clinical and molecular epidemiologic mechanisms of gastrointestinal neoplasms will improve risk stratification and enable clinicians to tailor their use of screening and surveillance strategies, chemopreventive agents, and therapeutic options. Progress in our understanding of the cellular and molecular pathways that underlie neoplastic transformation and tumor

References Dicken BJ, Bigam DL, Cass C, et al: Gastric adenocarcinoma: Review and considerations for future directions. Ann Surg 241:27-39, 2005. Houghton J, Wang TC: Helicobacter pylori infection and gastric cancer: A new paradigm for inflammation-associated epithelial cancers. Gastroenterology 1281567-1281578, 2005. Levin B, Lieberman DA, McFarland B, et al: Screening and surveillance for the early detection of colorectal cancer and adenomatous Polyps, 2008: A joint guideline

progression will provide additional targets for selective pharmacologic or immunologic therapies. Technologic advances will facilitate the endoscopic diagnosis and treatment of premalignant and malignant diseases of the gastrointestinal tract.

from the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology. CA Cancer J Clin 58:130-160, 2008. Shaheen N: Advances in Barrett esophagus and esophageal adenocarcinoma. Gastroenterology 128:1554-1566, 2005. Winawer SJ, Zauber AG, Fletcher RH, et al: Guidelines for colonoscopy surveillance after polypectomy: A consensus update by the US Multi-Society Task Force on Colorectal Cancer and the American Cancer Society. Gastroenterology 130:1872, 2006.

Chapter

VII

Diseases of the Pancreas David R. Lichtenstein

Anatomy and Physiology The pancreas is an organ located in the retroperitoneum (Fig. 40-1) that weighs between 70 and 120g and is about 12 to 20cm in length. The head of the pancreas is nestled in the C loop of the duodenum, and the tail extends obliquely posterior to the stomach toward the hilum of the spleen. The pancreas consists of the pancreatic acinus and islet cells. The acinar cells compose more than 95%, and the islets about 1% to 2%, of the pancreatic mass. Hormones that the islets produce include insulin, glucagon, somatostatin, and pancreatic polypeptide. The functional exocrine unit of the pancreas is the pancreatic acinus, which is composed of both acinar and ductal epithelial cells. Acinar cells synthesize proteolytic digestive enzymes, which are packaged separately in the Golgi region into condensing vacuoles and transported in an inactive form referred to as zymogens to the apical portions of the cell, where they are discharged into the central ductule of the acinus by exocytosis. The ductules coalesce to form larger ducts, which empty into the duodenum at the ampulla of Vater. Inactive enzymes secreted into the duodenum are converted to an active form by enterokinase secreted from small bowel enterocytes. Trypsinogen, converted to active trypsin in the duodenum by enterokinase, is the trigger enzyme that subsequently converts the other zymogens to active enzymes (Fig. 40-2). Enzymes secreted in an active form include lipase, amylase, and ribonuclease. The ductal cells secrete primarily water and electrolytes, which decrease the viscosity of the protein-rich acinar secretions and alkalinize gastric contents emptied into the duodenum to levels at which the pancreatic enzymes become catalytically active (pH ranges from >3.5 to 4).

Normal Pancreas Development At about 4 weeks of gestation, the dorsal pancreas forms as an evagination from the duodenum, and shortly thereafter, the ventral pancreas forms from the hepatic diverticulum. Rotation of the duodenum places the two pancreatic buds in close proximity at 7 to 8 weeks of gestation, at which time

their main ducts begin to fuse. If fusion is incomplete, the duct of Wirsung drains only the ventral pancreas through the major ampulla, and the duct of Santorini drains the bulk of the pancreas (dorsal pancreas) through the relatively small accessory ampulla. This common anomaly, termed pancreas divisum, is present in 5% to 10% of the general population and may be associated with acute and chronic pancreatitis. Theories suggest that pancreatitis may result from relative outflow obstruction of the main dorsal duct through the small accessory ampulla. Endoscopic papillotomy or surgical sphincteroplasty are two therapeutic maneuvers that may reduce the incidence of recurrent pancreatitis by increasing drainage through the accessory papilla.

Acute Pancreatitis Acute pancreatitis is best defined as an acute inflammatory process of the pancreas that may also involve peripancreatic tissues and remote organ systems. The overall incidence is 1 in 4000 for the general population. Most patients with acute pancreatitis have a mild course and recover with restoration of normal pancreatic function and gland architecture. However, in 10% to 20%, the various pathways that contribute to increased intrapancreatic and extrapancreatic inflammation result in what is generally termed systemic inflammatory response syndrome (SIRS). In some instances, SIRS predisposes to multiple organ dysfunction or pancreatic necrosis. Early steps in the management of patients with acute pancreatitis can decrease severity, morbidity, and mortality. Prevention of the septic and nonseptic complications in patients with severe acute pancreatitis depends largely on monitoring, vigorous hydration, and early recognition of pancreatic necrosis and choledocholithiasis.

CAUSES AND PATHOGENESIS The pathogenesis of acute pancreatitis remains incompletely understood. Based on experimental models, the initiating event in acute pancreatitis is intra-acinar activation of trypsin from trypsinogen, resulting in acute intracellular injury, pancreatic autodigestion, and the potential for 445

446

Section VII—Gastrointestinal Disease

Minor ampulla

Dorsal duct (Santorini)

Pylorus

Table 40-1 Causes of Acute Pancreatitis

Tail of pancreas

Obstructive Causes Gallstones Tumors: ampullary or pancreatic tumors Parasites: Ascaris or Clonorchis species Developmental anomalies: pancreas divisum, choledochocele, annular pancreas Periampullary duodenal diverticula Hypertensive sphincter of Oddi Afferent duodenal loop obstruction Toxins Ethyl alcohol Methyl alcohol Scorpion venom: excessive cholinergic stimulation causes salivation, sweating, dyspnea, and cardiac arrhythmias; seen mostly in the West Indies Organophosphorus insecticides Drugs

Major ampulla

Common Ventral duct Head of bile duct (Wirsung) pancreas Figure 40-1 Normal anatomy of the pancreas.

Definite association (documented with rechallenges): azathioprine/6-mercaptopurine, valproic acid, estrogens, tetracycline, metronidazole, nitrofurantoin, pentamidine, furosemide, sulfonamides, methyldopa, cytarabine, cimetidine, ranitidine, sulindac, dideoxycytidine Probable association: thiazides, ethacrynic acid, phenformin, procainamide, chlorthalidone, l-asparaginase Metabolic Causes Hypertriglyceridemia, hypercalcemia, end-stage renal disease

ENTEROCYTES

GUT LUMEN Trypsinogen Enterokinase

Trypsin

Trypsinogen Chymotrypsinogen Proelastase Procarboxypeptidases A and B Prophospholipase A2 Trypsin Chymotrypsin Elastase Carboxypeptidases A and B Phospholipase A2

Figure 40-2 Mechanism of proenzyme activation in the intestinal lumen. (Adapted from Solomon TE: Exocrine pancreas: Pancreatitis. In The Undergraduate Teaching Project in Gastroenterology and Liver Disease, Unit 24. Bethesda, Md, American Gastroenterological Association, 1984.)

Trauma Accidental: blunt trauma to the abdomen (car accident, bicycle) Iatrogenic: postoperative, endoscopic retrograde cholangiopancreatography, endoscopic sphincterotomy, sphincter of Oddi manometry Infectious Parasitic: ascariasis, clonorchiasis Viral: mumps, rubella, hepatitis A, hepatitis B, non-A and non-B hepatitis, coxsackievirus B, echovirus, adenovirus, cytomegalovirus, varicella virus, Epstein-Barr virus, human immunodeficiency virus Bacterial: mycoplasma, Campylobacter jejuni, tuberculosis, Legionella species, leptospirosis Vascular Ischemia: hypoperfusion (such as postcardiac surgery) or atherosclerotic emboli Vasculitis: systemic lupus erythematosus, polyarteritis nodosa, malignant hypertension Idiopathic

profound systemic complications once activated enzymes are leaked into the bloodstream. Initiating events may include obstruction of the pancreatic duct (e.g., gallstones, pancreatic tumor), overdistention of the pancreatic duct (e.g., from endoscopic retrograde cholangiopancreatography [ERCP]), reflux of biliary or duodenal juices into the pancreatic duct, changes in permeability of the pancreatic duct, ischemia of the organ, and toxin-induced cholinergic hyperstimulation. During the initial hospitalization for acute pancreatitis, reasonable attempts to determine etiology is appropriate, particularly those causes that may affect acute management. The cause of acute pancreatitis is readily identified in 70% to 90% of patients after an initial evaluation consisting

Ten to 30% of patients with pancreatitis; up to 60% of these patients have occult gallstone disease (biliary microlithiasis or gallbladder sludge). Other less common causes include sphincter of Oddi dysfunction and mutations in the cystic fibrosis transmembrane regulator. Miscellaneous Penetrating peptic ulcer Crohn disease of the duodenum Pregnancy associated Pediatric association: Reye syndrome, cystic fibrosis

of history, physical examination, focused laboratory testing, and routine radiologic evaluation. Gallstones account for 45%, alcohol 35%, miscellaneous causes 10%, and idiopathic causes 10% to 20% of acute pancreatitis cases (Table 40-1).

Chapter 40—Diseases of the Pancreas

CLINICAL MANIFESTATIONS Abdominal pain is virtually always present and may be severe and refractory to analgesics. Pain often radiates to the back and is usually worse when supine. The onset may be swift with pain reaching maximal intensity within 30 minutes, is frequently unbearable, and characteristically persists for more than 24 hours without relief. Physical examination usually reveals severe upper abdominal tenderness at times associated with guarding. Ileus occurs when the inflammatory process extends into the small intestinal and colonic mesentery or when a chemical peritonitis occurs. Other manifestations include nausea, vomiting, and fever caused by the significant inflammatory process and release of cytokines (Fig. 40-3). In acute pancreatitis, a wide variety of toxic materials, including pancreatic enzymes, vasoactive materials (e.g., kinins), and other toxic substances (e.g., elastase, phospholipase A2), are liberated by the pancreas and extravasate along fascial planes in the retroperitoneal space, lesser sac, and the peritoneal cavity. These materials cause chemical

irritation and contribute to third-space losses of protein-rich fluid, hypovolemia, and hypotension. These toxic materials may also reach the systemic circulation by lymphatic and venous pathways and contribute to subcutaneous fat necrosis and end-organ damage, including shock, renal failure, and respiratory insufficiency (atelectasis, effusions, and acute respiratory distress syndrome [ARDS]). Grey Turner sign (ecchymosis of the flank) or Cullen sign (ecchymosis in the periumbilical region) may be seen in association with hemorrhagic pancreatitis. Metabolic problems are common in severe disease and include hypocalcemia, hyperglycemia, and acidosis. Hypocalcemia is most commonly caused by concomitant hypoalbuminemia. Other mechanisms may include complexing of calcium to released free fatty acids, protease-induced degradation of circulating parathyroid hormone (PTH), and failure of PTH to release calcium from bone. Local spread of inflammation leads to effects on contiguous organs that include gastritis and duodenitis, splenic vein thrombosis, colonic necrosis, and external compression of the common bile duct, leading to biliary obstruction. Trypsin can activate

Progression

Death 11. Intractable shock

10. Adult respiratory distress syndrome

9. Vasodilation, vascular permeability, shock, acute renal failure

Bile reflux Ethanol Trauma Other causes

1. Damage to ductal epithelium

447

8. Activation of kallikrein system

7. Progression of injury (largely extrapancreatic in clinical manifestations)

4. Capillary and lymphatic injury 5. Capillary and lymphatic obstruction

6. Acinar cell injury and necrosis; release and activation of digestive enzymes and cell proteins

2. Leakage of 3. Activation of digestive juices proteolytic, lipolytic, or other enzymes in interstices of pancreas Figure 40-3 The pathophysiology of acute pancreatitis is not fully understood, but, as this schematic illustration implies, a cascade of events seems likely, beginning with the release of toxic substances into the parenchyma and ending with shock and death. Damage to the ductal epithelium or acinar cell injury may result from bile reflux, increased intraductal pressure, alcohol, or trauma. (Adapted from Grendell JH: The pancreas. In Smith LH Jr, Thier SO [eds]: Pathophysiology: The Biological Principles of Disease, 2nd ed. Philadelphia, WB Saunders, 1985, p 1228.)

448

Section VII—Gastrointestinal Disease

plasminogen to plasmin and induce clot lysis. On the other hand, trypsin can activate prothrombin and thrombin and produce thrombosis, leading to disseminated intravascular coagulation. Extrapancreatic fluid collections occur when fluid extravasates from the pancreas or surrounding leaky tissues. They are located in or near the pancreas and lack a wall of granulation or fibrous tissue. Acute fluid collections occur more commonly with severe pancreatitis. Most of these lesions regress spontaneously, and almost all remain sterile. The older term phlegmon was used in the past to describe inflammatory collections, but it is too ambiguous and imprecise for current use, given that it does not differentiate acute fluid collections from areas of pancreatic necrosis nor infected from noninfected collections. Pancreatic pseudocysts are defined as encapsulated nonepithelial lined collections of pancreatic juice, either pure or containing debris, single or multiple, small or large, and they can be located in or adjacent to the pancreas. Fluid collections must be present for a minimum of 4 weeks from the onset of pancreatitis to be termed a pseudocyst. Although most pseudocysts remain asymptomatic, presenting symptoms may include abdominal pain, early satiety, nausea, and vomiting due to compression of the stomach or gastric outlet. Rapidly enlarging pseudocysts may rupture, hemorrhage, obstruct the extrahepatic biliary tree, erode into surrounding structures, extend into the mediastinum, and become infected. Most pseudocysts less than 6cm in diameter will resolve over time, and one third of lesions less than 10cm in diameter remain asymptomatic or resolve. Indications for pseudocyst drainage include suspicion of infection or progressive enlargement with associated symptoms described previously. Asymptomatic pseudocysts should be followed. Pseudocysts can be drained surgically, percutaneously or endoscopically. The choice of treatment for symptomatic pseudocysts is frequently determined by the locally available expertise and by clinician preference because no method has been shown to be superior to the others. Pancreatic fistula occurs as a result of duct disruption and is treated with total parenteral nutrition, endoscopic stenting, and octreotide. Surgical intervention may be needed if the conservative approach is unsuccessful.

DIAGNOSIS The diagnosis of acute pancreatitis is based on a combination of clinical, biochemical, and radiologic factors. There is general acceptance that a diagnosis of acute pancreatitis requires two of the following three features: (1) abdominal pain characteristic of acute pancreatitis, (2) serum amylase or lipase more than 3 times the upper limit of normal, and (3) characteristic findings of acute pancreatitis on computed tomography (CT) scan. Elevated serum pancreatic enzymes may occur in a wide variety of other conditions, including bowel perforation, intestinal obstruction, mesenteric ische mia, tubo-ovarian disease, and renal failure. Serum lipase is slightly more specific and remains normal in some conditions associated with an elevation of serum amylase, including macroamylasemia, parotitis, and tubo-ovarian disease. The serum amylase level usually rises rapidly, as does the serum lipase level, and may remain elevated for 3 to 5 days. Serum lipase remains elevated longer than amylase and thus may be helpful if patients seek medical attention several days

after symptom onset. Repeated measurements of pancreatic enzymes have little value in assessing clinical progress, and the magnitude of serum amylase or lipase elevation does not correlate with the severity of pancreatitis. Macroamylase and macrolipase can occasionally cause isolated nonpathologic elevations of these enzymes, a situation in which the measurement of urinary clearance is useful. Pancreatic imaging with CT scanning can be used to confirm a diagnosis of pancreatitis (pancreatic enlargement, peripancreatic inflammatory changes, and extrapancreatic fluid collections). Selective CT scanning may also be useful in evaluating complications and assessing severity of disease (see later discussion), although a normal CT scan is present in 15% to 30% of mild cases. Acute gallstone pancreatitis should be suspected in patients with gallstones on ultrasonography or elevated liver tests, in particular an aspartate aminotransferase level elevated greater than threefold. Magnetic resonance imaging (MRI) is similar to CT with respect to imaging the inflamed pancreas and may be preferred in individuals at risk for contrast-induced injury (e.g., contrast allergy or renal insufficiency). However, recent studies also indicate the potential for gadolinium-induced nephrotoxicity with MRI examinations. MRI is also sensitive for the detection of necrosis, small neoplasms of the pancreas, and stones in the pancreaticobiliary tree.

SEVERITY OF DISEASE Supportive therapy alone is effective in treating 75% of all patients with acute pancreatitis. Twenty-five percent of patients, however, will suffer a complication, with one third succumbing to complications, yielding an overall mortality rate of 5% to 10%. Early deaths within the first 2 weeks are the result of multisystem organ failure caused by the release of inflammatory mediators and cytokines. Late deaths result from local or systemic infection. The risks for infection and death correlate with disease severity and the presence and extent of pancreatic necrosis. Therefore, a combination of clinical scoring and CT grade provides the most precise prognostic information. Patients should be stratified into mild or severe levels of illness based on well-established clinical criteria such as Ranson criteria (Table 40-2) or Acute Physiologic and Chronic Health Evaluation (APACHE II) scores. With increasing scores, the likelihood of a complicated, prolonged, and fatal outcome increases. The mortality rate is about 1% when fewer than three Ranson signs exist, 10% to 20% when three to five signs exist, and more than 50% when six Ranson signs exist. Similarly, an APACHE II score greater than 8 has been shown to predict severe pancreatitis. Conversely, a fatal outcome is unlikely with an APACHE II score less than 8. The distinction between interstitial and necrotizing acute pancreatitis has important prognostic implications (Fig. 40-4). Interstitial pancreatitis is characterized by an intact microcirculation and uniform enhancement of the gland on contrast-enhanced CT scanning. About 20% to 30% of patients with acute pancreatitis have necrotizing pancreatitis. Necrotizing pancreatitis is characterized by disruption of the pancreatic microcirculation so that large areas do not enhance on CT (Fig. 40-5). The presence of pancreatic necrosis predicts a worse severity of pancreatitis, particularly infection in the necrotic pancreatic tissue, also termed

Chapter 40—Diseases of the Pancreas Table 40-2 Signs Used to Assess Severity of Acute Pancreatitis At Time of Admission or Diagnosis Age > 55yr White blood cell count > 16,000/mm3 Blood glucose > 200mg/dL LDH > 2 × normal ALT > 6 × normal During Initial 48 Hours Decrease in hematocrit > 10% Serum calcium < 8mg/dL Increase in blood urea nitrogen > 5mg/dL Arterial Po2 < 60mmHg Base deficit > 4mEq/L Estimated fluid sequestration > 600mL ALT, alanine aminotransferase; LDH, lactate dehydrogenase; Po2, partial pressure of oxygen. Data from Ranson JH, Rifkind KM, Turner JW: Prognostic signs and nonoperative peritoneal lavage in acute pancreatitis. Surg Gynecol Obstet 43:209-219, 1976. By permission of Surgery, Gynecology and Obstetrics.

449

infected necrosis. Infected necrosis develops in 30% to 50% of patients with acute necrotizing pancreatitis but rarely in those with interstitial disease ( 7 g per 24 hours); however, the test is not specific for pancreatic exocrine insufficiency. The test also lacks sensitivity because steatorrhea will not occur in chronic pancreatitis until pancreatic lipase output falls to less than 5% to 10% of normal. The serum trypsinogen level correlates with functioning acinar parenchyma. A low level (2.5 to 3 mg/dL) results in jaundice and defines icteric hepatitis. Values higher than 20mg/dL are uncommon and approximately correlate with the severity of disease. Elevations in serum alkaline phosphatase are usually limited to 3 times normal levels, except in cases of cholestatic hepatitis. A complete blood cell count most commonly shows mild leukopenia with atypical lymphocytes. Anemia and thrombocytopenia may also be present. The icteric phase of acute viral hepatitis may last days to weeks, followed by gradual resolution of symptoms and laboratory values.

SERODIAGNOSIS 1

6 5 Months

C Figure 43-2 Sequence of clinical and laboratory findings in (A) a patient with acute hepatitis A virus (HAV) infection, (B) a patient with hepatitis B virus (HBV) infection, and (C) a patient with hepatitis C virus (HCV) infection. ALT, alanine transaminase; HBc, hepatitis B core; HBe, hepatitis B early; HBeAg, hepatitis B early antigen; HBs, hepatitis B surface; HBsAg, hepatitis B surface antigen; IgG, immunoglobulin G; PCR, polymerase chain reaction.

risk, intravenous drug abusers, infants of infected mothers (vertical transmission), and health professionals. Patients with increased exposure to blood or blood products or with impaired immunity (e.g., patients undergoing dialysis and patients with leukemia, hemophilia, or trisomy 21 syndrome) are also highly susceptible to HBV. HCV was the main cause of post-transfusion hepatitis before 1992. It is presently the most common cause of hepatitis in intravenous drug users, and it accounts for a substan-

The ability to detect the presence of viral nucleic acids in hepatitis B, C, and D and antigen or antibodies to components of hepatitis A through E has fostered progress in the epidemiology of viral hepatitis. These viral markers are used in the diagnosis of acute viral hepatitis (see Fig. 43-2; Tables 43-2 and 43-3). An etiologic diagnosis is of great importance in planning preventive and public health measures pertinent to the close contacts of infected patients and in evaluating prognosis. Epstein-Barr virus and cytomegalovirus hepatitis may also be diagnosed by the appearance of specific antibodies of the immunoglobulin M (IgM) class. In acute hepatitis B, HBsAg and HBeAg are present in serum. Both are usually cleared within 3 months, but HBsAg may persist in some patients with uncomplicated cases for 6 months to 1 year. Clearance of HBsAg is followed after a variable “window” period by emergence of anti-HBs, which confers long-term immunity. Anti-HBc and anti-HBe appear in the acute phase of the illness, but neither provides immunity. Uncommonly, during the serologic window period, anti-HBc IgM, a marker of active viral replication suggesting recent infection, may be the only evidence of HBV infection. HDV infec-

Chapter 43—Acute and Chronic Hepatitis

469

Table 43-2 Serologic Markers of Viral Hepatitis Agent

Marker

Definition

Significance

Hepatitis A virus (HAV)

Anti-HAV IgM type IgG type

Antibody to HAV — —

Hepatitis B virus (HBV)

HBsAg

HBV surface antigen

HBeAg

HBe antigen; a component of the HBV core

— Current or recent infection or convalescence Current or previous infection; conferring immunity Positive in most cases of acute or chronic infection Transiently positive in acute hepatitis B

HBV DNA Anti-HBe

Anti-HBc (IgM or IgG)

Anti-HBs Hepatitis C virus (HCV)

Hepatitis D virus (HDV)

Antibody to HBV core antigen

Positive in late convalescence in most acute cases Anti-HCV

Antibody to HBV surface antigen and after vaccination Antibodies to a group of recombinant HCV peptides

HCV RNA

Infectious viral genomic material Antibody to HDV antigen Viral peptide Infectious viral genomic material

Anti-HDV (IgM or IgG) HDV antigen HDV RNA

Hepatitis E virus (HEV)

Infectious viral genomic material Antibody to HBe antigen

Anti-HEV (IgM or IgG)

Antibody to HEV antigen

May persist in chronic infection Reflection of presence of viral replication, whole Dane particles in serum, high infectivity Serum level reflects degree of viral replication; predicts response to therapy Transiently positive in convalescence Persistently present in some chronic cases Usually a reflection of low infectivity Positive in all acute and chronic cases Reliable marker of infection, past or current IgM anti-HBc a reflection of active viral replication and acute infection Not protective Confers immunity Positive on average 12wk after exposure; not protective Persistent in acute, chronic, or past infection Reflects ongoing infection, level inversely linked to treatment response Acute or chronic infection seen with positive HBsAg; not protective IgM and IgG clear in resolving infection IgG persists in chronic infection Persists in chronic infection Most reliable test for acute or chronic infection Acute or chronic infection IgM may persist up to 6 months

IgG, immunoglobulin G; IgM, immunoglobulin M.

tion superimposed on HBV infection is most reliably detected by polymerase chain reaction (PCR) testing for HDV RNA. Other possible tests include HDV antigen and anti-HDV (IgM and IgG antibodies). Acute hepatitis C can be detected using a sensitive PCR assay for HCV RNA within 2 weeks of exposure. Serum antibodies to HCV develop within 12 weeks of exposure or within 4 to 5 weeks after biochemical abnormalities are discovered. At onset of symptoms, 30% of patients will be missed if checked by serum enzyme immunoassay (EIA) for HCV antibody alone. Commercial EIAs for hepatitis E to detect both IgM and IgG class antibodies are also available but may lack general sensitivity and specificity. The HEV IgM antibody is present for up to 6 months after exposure.

COMPLICATIONS Cholestatic Hepatitis In some patients, most commonly during HAV infection, a self-limited period of cholestatic jaundice may supervene that is characterized by marked conjugated hyperbilirubinemia, elevation of alkaline phosphatase, and pruritus. Investigation may be required to differentiate this condition from mechanical obstruction of the biliary tree (see Chapter 45).

Fulminant Hepatitis Massive hepatic necrosis occurs in less than 1% of patients with acute viral hepatitis and leads to a devastating and often

470

Section VIII—Diseases of the Liver and Biliary System

Table 43-3 Interpretation of Serologic Markers and Serum DNA in Hepatitis B HBsAg HBeAg Acute hepatitis Acute hepatitis, window period Recovery from acute hepatitis Chronic hepatitis Chronic hepatitis (precore mutant) Inactive carrier Vaccinated

+/−

+ +

Anti-HBc IgM Anti HBc IgG Anti-HBs Anti-HBe HBV DNA* + + +

+/− +

+ +

+/− +

HBsAg, hepatitis B surface antigen; HBeAg, hepatitis Be antigen; anti-HBc IgM, hepatitis B core antibody (IgM type); anti-HBc IgG, hepatitis B core antibody (IgG type); anti-HBs, hepatitis B surface antibody; anti-HBe, hepatitis Be antibody; HBV DNA, hepatitis B viral DNA. *HBV DNA > 105 copies/mL.

fatal condition called fulminant hepatic failure. This condition is discussed in detail in Chapter 44.

Chronic Hepatitis Hepatitis A does not progress to chronic liver disease, although occasionally it has a relapsing course. Persistence of aminotransferase elevation and viral antigens or nucleic acids beyond 6 months in patients with hepatitis B and C suggests evolution to chronic hepatitis, although slowly resolving acute hepatitis may occasionally lead to such test abnormalities for up to 12 months, with eventual complete resolution. Chronic hepatitis is considered in detail later in this chapter.

Rare Complications Acute viral hepatitis may be followed by aplastic anemia, which affects mostly male patients and results in a mortality rate of greater than 80%. Pancreatitis, myocarditis, pericarditis, pleural effusion, and neurologic complications, including Guillain-Barré syndrome, aseptic meningitis, and encephalitis, have also been reported. Cryoglobulinemia and glomerulonephritis are associated with hepatitis B and C, and polyarteritis nodosa with hepatitis B.

MANAGEMENT All cases of acute hepatitis A, B, and E, unless complicated by fulminant hepatitis, are self-limited (see Table 43-2). Treatment of acute hepatitis C is important, and recent data support that early treatment within 12 weeks of diagnosis with pegylated interferon-α induces high sustained virologic response rates. Studies of antiviral therapy in acute hepatitis B have not shown clear benefit, although many experts advocate use of nucleoside or nucleotide analogues, specifically in the setting of acute liver failure due to hepatitis B. The treatment in all other cases is largely supportive and includes rest, maintenance of hydration, and adequate dietary intake. Most patients show a preference for a low-fat, high-carbohydrate diet. Alcohol should be avoided. Vitamin supplementation is of no proven value, although vitamin K may be indicated if prolonged cholestasis occurs. Nausea can be treated with small doses of metoclopramide and hydroxyzine. Hospitalization is indicated for patients with severe nausea and vomiting and for those with evidence of deteriorating liver function, such as hepatic encephalopathy or pro-

longation of the prothrombin time. In general, hepatitis A and E may be regarded as noninfectious after 3 weeks, whereas hepatitis B is potentially infectious to sexual contacts throughout its course, although the risk is low once HBsAg has cleared.

PREVENTION Both feces and blood from patients with hepatitis A and E contain virus during the prodromal and early icteric phases of the disease (see Fig. 43-2). Raw shellfish concentrate the HAV from sewage pollution and may serve as vector of the disease. General hygienic measures should include hand washing by contacts and careful handling, disposal, and sterilization of excreta and contaminated clothing and utensils. Close contacts of patients with hepatitis A should receive anti-HAV serum immunoglobulin as soon as possible after exposure. HAV vaccination is appropriate for children and travelers to endemic areas, individuals with immunodeficiency or chronic liver disease, and those with high-risk behaviors or occupations. Recently, a recombinant vaccine for HEV has been shown to be safe and effective in a highrisk population (healthy adults in an endemic area) in a randomized controlled trial, but formal guidelines for its use have not been established. Hepatitis B is rarely transmitted by body fluids other than blood. However, it is highly infectious, and strict adherence to universal precautions is mandatory. Efforts at preventing hepatitis B have involved the use of immunoglobulin-enriched anti-HBs (HBIG) and recombinant HBV vaccines. Postexposure prophylaxis with HBIG after blood or mucosal exposure (e.g., needlestick, eye splash, sexual contacts of patients with acute hepatitis B, neonates born to mothers with acute or chronic infection) should be given within 7 days along with HBV vaccine. Preventive vaccination is currently recommended for highrisk groups and individuals (health care professionals, patients undergoing dialysis, patients with advanced liver disease or hemophilia, residents and staff of custodial care institutions, sexually active homosexual men) and is advocated universally for children. No accepted prevention strategies are available for HCV. Serum immunoglobulin is not useful for postexposure prophylaxis. The advent of widespread blood product screening for anti-HCV has made post-transfusion hepatitis a rarity.

Chapter 43—Acute and Chronic Hepatitis

Alcoholic Fatty Liver and Hepatitis Alcohol abuse is a major cause of liver disease in the Western world. Three major pathologic lesions resulting from alcohol abuse are (1) fatty liver, (2) alcoholic hepatitis, and (3) cirrhosis. These lesions are not mutually exclusive, and there may be features of all three lesions in the same patient. The first two lesions are potentially reversible and may sometimes be confused clinically with viral hepatitis or gallbladder or biliary tract disease. Alcoholic cirrhosis is discussed in Chapter 45.

MECHANISM OF INJURY Mechanisms of liver injury caused by alcohol are complex. Ethanol and its metabolites, acetaldehyde and nicotinamide adenine dinucleotide phosphate (NADP), are directly hepatotoxic and cause a large number of metabolic derangements. Induction of cytochrome P-450 (CYP2E1) and cytokine pathways, particularly tumor necrosis factor-α (TNF-α), is also critical in initiating and perpetuating hepatic injury and producing the lesions of alcoholic hepatitis. Hepatotoxic effects from alcohol vary considerably among individuals. Nevertheless, consumption by men of 40 to 80g of ethanol per day (one beer or one mixed drink = 10g of ethanol) for 10 to 15 years carries a substantial risk for the development of alcoholic liver disease, whereas women appear to have a lower threshold of injury. Malnutrition and presence of other forms of chronic liver disease may potentiate the toxic effects of alcohol on the liver, and genetic factors may contribute to individual susceptibility.

CLINICAL AND PATHOLOGIC FEATURES Alcoholic fatty liver may exhibit as incidentally discovered tender hepatomegaly. Some patients consult a physician because of pain in their right upper quadrant. Jaundice is rare. Aminotransferases are mildly elevated (10 to 15g), the formation of excess toxic metabolites depletes the available glutathione and produces necrosis. Acetaminophen overdose, commonly taken in a suicide attempt, leads to nausea and vomiting within a few hours. These symptoms subside and are followed in 24 to 48 hours by clinical and laboratory evidence of hepatocellular necrosis (raised aminotransferase levels) and hepatic dysfunction (prolonged prothrombin time and hepatic encephalopathy). Similar findings may occur with therapeutic doses of acetaminophen in patients with chronic alcoholism or malnutrition. Extensive liver necrosis may lead to fulminant hepatic failure and death. In a patient with nonstaggered overdose, a serum acetaminophen level should be drawn 4 to 24 hours after ingestion. If plotted on a treatment nomogram of plasma drug concentration against time, it can predict the severity of outcome and need for therapy. Treatment with N-acetylcysteine given orally (140-mg/kg bolus followed by 70mg/kg × 17 doses) or intravenously, thought to promote hepatic glutathione synthesis, may be life saving. Nonsteroidal anti-inflammatory drugs (NSAIDs) as a class are an important cause of drug-induced liver disease. Salicylates cause dose-dependent hepatocellular injury that is usually clinically mild and easily reversible. Diclofenac, one of the most commonly prescribed NSAIDs worldwide, has been linked to asymptomatic elevation of aminotransferases, acute hepatitis, and fulminant hepatic failure. Sulindac is considered the most likely NSAID to produce hepatic injury and causes a damage spectrum ranging from hepatocellular to mixed to pure cholestatic injury. Whether the newer cyclo-oxygenase-2–selective NSAIDs have a lower risk for hepatotoxicity is uncertain.

ANTIBIOTICS AND ANTIVIRALS Antimicrobials as a class are the most frequently incriminated agents causing drug-induced liver injury. Amoxicillin–clavulanic acid is the leading cause of antibiotic-related liver injury and results in a cholestatic hepatitis. Men appear to be more susceptible than women. Other common agents include nitrofurantoin, isoniazid, trimethoprim-sulfamethoxazole, and fluoroquinolones. Isoniazid, as a singledrug prophylaxis against tuberculosis, commonly produces raised serum aminotransferase levels in 20% of patients. This effect appears to be transient and self-limited in most patients. However, a 1% incidence exists of clinical hepatitis,

Chapter 43—Acute and Chronic Hepatitis which progresses to fatal hepatic necrosis in 10% of affected patients. Individual and age-related differences in hepatic acetylation of potentially toxic isoniazid metabolites may be important in this injury. Thus, the incidence of severe hepatic damage increases with age such that significant elevation of aminotransferase levels in persons who are older than 35 years is an indication for discontinuing the drug. Erythromycin is an established agent causing cholestatic injury. Trimethoprim-sulfamethoxazole characteristically causes cholestatic or mixed injury. A large number of agents used to treat HIV infection have been linked with hepatic injury of various forms. Important among these agents are nevirapine, ritonavir, and indinavir.

CENTRAL NERVOUS SYSTEM AGENTS Central nervous system agents are second only to antimicrobials as a frequent cause of drug-induced liver injury. Important categories are anticonvulsants and anesthetics. Other common agents include duloxetine, bupropion, and fluoxetine. Among anticonvulsants, sodium valproate, phenytoin, carbamazepine, and lamotrigine are the most common drugs causing liver injury. Phenytoin and carbamazepine have been implicated in an antiepileptic hypersensitivity syndrome, characterized by a triad of rash, fever, and hepatocellular injury that may lead to fulminant hepatic failure. Lymphadenopathy and a mononucleosis-like picture with atypical lymphocytes may be present. Renal and pulmonary involvement may also occur. Historically, the anesthetic agent halothane caused an uncommon acute viral hepatitis– like reaction several days after exposure in susceptible persons. Hepatic injury was caused in part by an allergic response to hepatic neoantigens produced by halothane metabolism, and the severity of this reaction increased with repeated exposure. Newer, commonly used halogenated anesthetic agents (e.g., isoflurane, enflurane) are hepatotoxic in a much smaller number of patients, though cross-sensitivity does exist.

HERBS Herbal supplements are taken throughout the world, and about $5 billion per year is spent in the United States alone on herbal agents. Incorrectly considered to be safe because they are natural, many herbs are hepatotoxic. Senecio, Heliotropium, Crotalaria, and comfrey contain pyrrolizidine alkaloids that cause hepatic veno-occlusive disease. Hepatotoxicity ranging from mild hepatitis to massive necrosis and fulminant hepatic failure has been associated with the use of chaparral, germander, pennyroyal oil, mistletoe, valerian root, comfrey, and Ma huang. Milk thistle, often taken by patients with chronic hepatitis and cirrhosis, has not been associated with hepatotoxicity, but its benefit is undefined because of a lack of controlled studies.

Chronic Hepatitis Chronic hepatitis is defined as a hepatic inflammatory process that fails to resolve after 6 months and, in those with acute viral hepatitis, by persistence of serum viral antigens and nucleic acids beyond a similar period.

473

ETIOLOGY Acute viral hepatitis can ultimately lead to chronic hepatitis, with the notable exceptions of HAV and HEV. Nonalcoholic steatohepatitis (NASH) is now considered the most frequent cause of chronic hepatitis in the United States and Western Europe. Several drugs may produce chronic hepatitis, the best recognized being methyldopa. In contrast to acute hepatitis, an etiologic agent is sometimes difficult to identify in cases of chronic hepatitis. The pathogenesis of these idiopathic forms may represent quiescent autoimmune disease, undetected past drug-induced injury or NASH, antibodynegative viral infections, or misdiagnosed cholestatic liver injury (e.g., primary biliary cirrhosis, primary sclerosing cholangitis).

CLASSIFICATION Current classification of chronic hepatitis is based on the etiologic agent responsible for disease, the grade of injury (determined by the numbers and location of inflammatory cells), and the stage of disease on liver biopsy (determined by the degree, location, and distortion of normal architecture by fibrosis). This classification allows integration of knowledge of the natural history of specific causes with histologic features of hepatic damage to assess the severity and prognosis of the process. Thus, in general, biochemical and serologic studies along with liver biopsy are used in the diagnosis and management of chronic hepatitis.

Chronic Viral Hepatitis Chronic hepatitis B follows acute hepatitis B in 5% to 10% of adults in the United States. HBV infection without evidence of any liver damage may persist, resulting in asymptomatic or healthy hepatitis B carriers. In Asia and Africa, many such carriers appear to have acquired the virus from infected mothers during infancy (vertical transmission). Patients who are HBsAg and HBeAg positive and have high serum HBV DNA (>20,000 IU/mL or >100,000 copies/mL), coupled with increased serum aminotransferases (ALT, 2 times normal) are in a high replicative phase (see Table 43-3). In contrast, patients in a low replicative phase are HBsAg and anti-HBe positive, have low serum HBV DNA (10mmHg). The HVPG is measured using a transvenous approach by subtracting free hepatic venous pressure from the wedged hepatic venous pressure. Although cirrhosis is the most important cause of portal hypertension, any process leading to increased resistance to portal blood flow into (presinusoidal) or through (sinusoidal) the liver or to hepatic venous outflow from the liver (postsinusoidal) may result in portal hypertension (Table 45-3). In addition, cirrhosis is associated with increased cardiac output, which leads to greater splanchnic blood flow, further aggravating portal hypertension. It is important to recognize that the HVPG is reliably increased only in sinusoidal portal hypertension. In an attempt to decompress the portal system, venous collaterals form between the portal and systemic circulations. Major sites of collateral vessel formation include the gastroesophageal junction, retroperitoneum, rectum, and falciform ligament of the liver (abdominal and periumbilical collaterals). Clinically, the most important collaterals are those connecting the portal vein to the azygos vein through dilated, tortuous veins (varices) in the submucosa of the gastric fundus and esophagus.

Table 45-3 Causes of Portal Hypertension Increased Resistance to Flow Presinusoidal Extrahepatic • Portal or splenic vein occlusion (thrombosis, sclerosis, tumor) Intrahepatic • Schistosomiasis • Congenital hepatic fibrosis • Sarcoidosis Sinusoidal Cirrhosis (many causes) Alcoholic hepatitis Postsinusoidal Intrahepatic • Veno-occlusive disease Extrahepatic • Budd-Chiari syndrome • Cardiac causes: constrictive pericarditis Increased Portal Blood Flow Splenomegaly not caused by liver disease Arterioportal fistula

Variceal Hemorrhage Gastroesophageal varices may develop when the portal pressure gradient exceeds 10 mm Hg, and the risk for variceal rupture leading to hemorrhage occurs when the gradient is greater than 12 mm Hg. Hemorrhage develops in 10% to 30% of patients every year, and each episode of variceal hemorrhage is associated with a mortality rate as high as

Chapter 45—Cirrhosis of the Liver and Its Complications Primary Prevention (nonselective β blockers and/or mononitrates)

Treating first variceal hemorrhage

1. Somatostatin or analogues 2. Endoscopic therapy 3. Antibiotic prophylaxis

Successful

Unsuccessful

Secondary prevention 1. Endoscopic ablation therapy and 2. Nonselective β blockers

1. Sengstaken-Blakemore tube 2. Portal systemic shunt surgery Portacaval anastomosis TIPS Distal splenorenal shunt

Consider for liver transplantation Figure 45-2 Prevention and treatment of variceal bleeding. TIPS, transjugular intrahepatic portosystemic shunt.

15% to 30%. Bleeding occurs most commonly from large varices in the esophagus when high tension in the walls of these vessels leads to rupture. Among gastric varices, fundal varices have the highest rate of bleeding and may bleed with portal pressure gradients of less than 12mmHg. Variceal bleeding usually causes painless hematemesis, melena, or hematochezia, which typically leads to hemodynamic compromise (see Chapter 34), further aggravated by impaired hepatic synthesis of coagulation factors (from hepatocellular dysfunction) and thrombocytopenia (from hypersplenism). The management of gastroesophageal varices includes the treatment of acute variceal hemorrhage, the prevention of rebleeding (secondary prophylaxis), and the prevention of the initial episode of bleeding (primary prophylaxis) (Fig. 45-2). In the setting of acute variceal hemorrhage, the initial intervention consists of hemodynamic resuscitation with colloids such as blood or fresh-frozen plasma and airway protection and ventilatory support, if necessary. Combined pharmacologic and endoscopic therapy is superior to either therapy alone, especially if pharmacologic therapy is instituted immediately. In addition, prophylactic intravenous antibiotics should be administered early because they reduce the risk for infection, rebleeding, and death. Current pharmacologic therapy consists of somatostatin or its synthetic analogues (i.e., octreotide, vapreotide). These agents are best instituted before endoscopic examination. Endoscopic therapy includes band ligation and/or sclerotherapy (Web Video 45-1). Prospective studies have demonstrated that band ligation is the preferred modality given the lower inci-

481

dence of adverse effects and complications. In patients with gastric variceal hemorrhage, endoscopic variceal ablation with cyanoacrylate glue is superior to band ligation, although this therapy is not approved in the United States. Balloon tamponade (Sengstaken-Blakemore tube, Linton tube, or Minnesota tube) is a temporary measure reserved for patients in whom endoscopic therapy fails in the setting of massive hemorrhage. These patients may need to undergo portal decompression through a surgical shunt or transjugular intrahepatic portosystemic shunt (TIPS) placement. After an initial episode of variceal bleeding, recommendations for secondary prophylaxis include a combination of nonselective β blockers (propranolol and nadolol) and variceal obliteration through repeated courses of endoscopic band ligation. However, individual patient characteristics may dictate whether nonselective β blockers can be used. There are no formal recommendations on frequency and duration of follow-up endoscopy, although it is generally done at 1- to 6-month intervals after obliteration of varices. Endoscopic screening has been recommended to identify patients at high risk for variceal bleeding (i.e., those with large varices) so that primary prophylaxis can be instituted. Several studies suggest that certain clinical features (e.g., thrombocytopenia, ascites, telangiectasias) may help predict patients who are likely to have large varices, but given the poor predictive values of these features, endoscopic screening should be performed in all patients newly diagnosed with cirrhosis. Capsule endoscopy has been evaluated as a less invasive screening alternative to upper endoscopy but is currently less sensitive in diagnosing varices. Primary prophylaxis should be instituted for all patients with large varices and in those with advanced liver disease (Child class B and C) with small varices. Nonselective β blockers are the agents of choice for primary prophylaxis because they reduce portal blood flow and vascular resistance and hence portal pressure. In patients with contraindications or intolerance to β blockers, variceal obliteration through endoscopic band ligation is the best alternative.

Ascites Ascites is the accumulation of excess fluid in the peritoneal cavity. Although cirrhosis is the most common cause of ascites, this condition may have numerous other causes (Table 45-4). The serum ascites-albumin gradient has replaced the exudative-transudative classification of ascites. An elevated serum ascites-albumin gradient (>1.1 g/dL, serum albumin concentration–ascites albumin concentration) signifies the presence of portal hypertension. Ascites becomes clinically detectable with fluid accumulation greater than 500 mL. Shifting dullness to percussion is the most sensitive clinical sign of ascites, but ultrasonography can readily detect smaller fluid volumes (250mL). The precise sequence of events leading to the development of cirrhotic ascites remains debated. However, both excess renal sodium and water retention resulting from portal hypertension and splanchnic vasodilation resulting in overflow of fluid into the peritoneum (overflow theory) and decreased effective circulating blood volume resulting from systemic arterial vasodilation leading to activation of

482

Section VIII—Diseases of the Liver and Biliary System

neurohumoral systems and sodium and water retention (underflow theory) play a role. Management of cirrhotic ascites consists initially of sodium restriction, preferably to less than 2 g per day. Restricted fluid intake may be necessary if hyponatremia (1.1g/dL

Low: λ) γ, α, or µ heavy chain or fragment Free light chain (λ > κ) IgG > IgM > IgA, usually without urinary light-chain secretion M protein occasionally secreted; IgM > IgG M protein occasionally secreted; IgM > IgG

Nonlymphoid Neoplasms Chronic myelogenous leukemia Carcinomas (e.g., colon, breast, prostate)

No consistent patterns No consistent patterns

Autoimmune or Autoreactive Disorders Cold agglutinin disease Mixed cryoglobulinemia Sjögren syndrome

IgM κ most common IgM or IgA IgM

Anti-I antigen Anti-IgG

Miscellaneous Inflammatory, Storage, or Infectious Disorders Lichen myxedematosus Gaucher disease Cirrhosis, sarcoid, parasitic diseases, renal acidosis

IgG λ IgG No consistent pattern

IgA, immunoglobulin A; IgD, immunoglobulin D; IgG, immunoglobulin G; IgM, immunoglobulin M. Modified from Salmon SE: Plasma cell disorders. In Wyngaarden JB, Smith LH Jr (eds): Cecil Textbook of Medicine, 18th ed. Philadelphia, WB Saunders, 1988, p 1026.

Chapter 51—Disorders of Lymphocytes occurs in about 1% of patients per year. Distinguishing patients with stable, nonprogressive MGUS from patients in whom multiple myeloma will eventually develop is difficult. The risk for progression is greater in patients with IgA or IgM M proteins and in patients with initial concentrations of M protein in excess of 1.5 g/dL. Although no definitive evidence has been found that monitoring patients with the diagnosis of MGUS improves survival, patients should undergo annual evaluation, including serum electrophoresis, to detect progression to multiple myeloma before the onset of overt symptoms or complications.

MULTIPLE MYELOMA Multiple myeloma is a malignant plasma cell disorder characterized by neoplastic infiltration of the bone marrow and bone and the presence of monoclonal immunoglobulin or light chains in the serum or urine. The diagnosis of multiple myeloma is made by identifying an increase in the number of plasma cells in the bone marrow (>30%) and a serum M protein other than IgM exceeding 3g/dL for IgG or 2g/ dL for IgA or a urine M protein exceeding 1g per 24 hours. Patients with lower levels of M protein or less than 30% bone marrow plasmacytosis may still be diagnosed with myeloma based on the presence of a combination of other features such as hypogammaglobulinemia, lytic bone lesions, or plasmacytoma. For patients lacking these features, the major differential diagnosis is usually between MGUS and myeloma; in some cases, the distinction can only be made by serial follow-up of the patient with evidence of rising M protein levels or the development of associated clinical manifestations of myeloma. About 20% of patients with multiple myeloma do not have detectable serum M protein by standard electrophoresis but have circulating free light chains that appear in the urine (Bence Jones protein) that can be detected in a 24-hour urine collection by urine protein electrophoresis (light-chain disease). In rare cases, patients with nonsecretory myeloma have neither detectable serum nor urine M protein. However, in these patients, a monoclonal population of plasma cells can be detected by immunohistochemical identification of cytoplasmic lightchain–restricted immunoglobulin. Quantitative assays for detection of free light chains in the serum of patients with multiple myeloma have now become widely available and may be used to assess disease in a similar fashion to electrophoretic measurements. These assays are quite sensitive and may provide measurement of clonal protein in patients thought to have nonsecretory disease by other methods. Free light chains have a relatively short half-life in the circulation, 2 to 6 hours, in comparison with weeks for intact immunoglobulin molecules and may therefore be used to obtain a more rapid assessment of disease response for patients on therapy. The clinical manifestations of multiple myeloma relate to the direct effects of bone marrow and bone infiltration by malignant plasma cells, the systemic effects of the M protein, or the effects of the concomitant deficiency in humoral immunity that occurs in this disease. The most common symptom in multiple myeloma is bone pain. Bone radiographs typically show pure osteolytic punched-out lesions, often in association with generalized osteopenia and pathologic fractures. Bony lesions can show as expansile masses

551

associated with spinal cord compression. Hypercalcemia caused by extensive bony involvement is common in myeloma and may dominate the clinical picture. Anemia occurs in most patients as a result of marrow infiltration and suppression of hematopoiesis and results in fatigue; granulocytopenia and thrombocytopenia are less common. Patients with myeloma are susceptible to bacterial infections because of impaired production and increased catabolism of normal immunoglobulins. Respiratory tract infections from Streptococcus pneumoniae, Staphylococcus aureus, H. influenzae, and Klebsiella pneumoniae, and gram-negative urinary tract infections are common. Renal insufficiency occurs in about 25% of patients with myeloma. The cause of renal failure in these patients is often multifactorial; hypercalcemia, hyperuricemia, infection, and amyloid deposition can contribute. However, direct tubular damage from light-chain excretion is invariably present. M proteins can also cause a host of diverse effects because of their physicochemical properties. These effects include cryoglobulinemia, hyperviscosity, amyloidosis, and clotting abnormalities resulting from interaction of the M protein with platelets or clotting factors. Several staging or classification systems exist for myeloma. The three-tier staging system for myeloma is a functional system that correlates with survival (Table 51-7). In contrast to the anatomic staging systems used for lymphomas and solid tumors, myeloma staging is based on clinical tests (bone radiographs) and laboratory tests (hemoglobin, serum calcium, serum or urine M protein levels, and serum creatinine) that correlate with tumor burden. Adverse prognostic factors include advanced stage, impaired renal function, elevated LDH levels, abnormal bone marrow cytogenetics, depressed serum albumin levels, and elevated β2-microglobulin levels. The last is the single most powerful predictor of survival. Recently, a simplified prognostic scheme, the International Staging System for Myeloma, identified three stages with distinct prognosis based on only two variables: β2-microglobulin and albumin levels. A classification system proposed by the International Myeloma

Table 51-7 Myeloma Staging System Stage

Criteria

All of the following: 1. Hemoglobin >10g/dL 2. Serum calcium 6mo Lifelong

*Long-term therapy must be adjusted individually according to other diseases, risks for bleeding, presence of transient risk factors, and ease of compliance. † Inherited risk factors include factor V Leiden; prothrombin 20210A; deficiencies of antithrombin III, protein C, or protein S. VTE/PE, venous thromboembolism/pulmonary embolism.

Table 54-8 Drugs that Affect Warfarin Levels Drugs that Increase Warfarin Levels: Prolonged INR ↓ Warfarin clearance Disulfiram Metronidazole Trimethoprim-sulfamethoxazole ↓ Warfarin-protein binding Phenylbutazone ↑ Vitamin K turnover Clofibrate Drugs that Decrease Warfarin Levels: Subtherapeutic INR ↑ Hepatic metabolism of warfarin Barbiturates Rifampin ↓ Warfarin absorption Cholestyramine ↑, Increased; ↓, decreased; INR, international normalized ratio.

warfarin is a teratogen; effective contraception should be used concurrently in women of childbearing age. Supratherapeutic INR levels commonly occur with warfarin therapy, with or without bleeding. In patients with moderately elevated INR values (>5) with little or no bleeding, temporary discontinuation of warfarin and reinstitution of the drug at a lower maintenance dose may be sufficient. Patients with higher INR values (5 to 9) without serious bleeding should have warfarin withheld and receive low doses (1 to 2.5mg per day) of oral vitamin K to reach therapeutic INR levels; parenteral vitamin K can be given if gastrointestinal function is problematic. When serious active bleeding occurs with high INR values, especially if surgery is required to correct the bleeding, a combination of vitamin K and transfusion of plasma (see Chapter 53) will rapidly correct the INR. The INR can become elevated as a result of concurrent use of drugs that increase free warfarin levels (Table 54-8). Whenever bleeding occurs as a complication of anticoagulation, serious consideration must be given to

590

Section IX—Hematologic Disease

future bleeding risks and to whether the patient requires prophylactic IVC filter placement.

Antithrombotic Therapy during Pregnancy Heparins, both UFH and LMWH, are the safest therapy for venous thrombosis during pregnancy; heparin does not cross the placenta, unlike warfarin, which causes a characteristic fetal embryopathy. Warfarin also causes fetal hemorrhage and placental abruption and should be avoided during pregnancy. VTE or PE during pregnancy should be treated with intravenous UFH for 5 to 10 days, followed by an adjusted-dose regimen of subcutaneous UFH, starting with 20,000 U every 12 hours and adjusted to achieve a PTT higher than 1.5 times baseline at 6 hours after injection. An attractive alternative to UFH during pregnancy is LMWH, which can be given subcutaneously once or twice daily and does not require monitoring. Suprarenal IVC filters have also been used successfully during pregnancy without significant morbidity. In women with aPL syndrome who become pregnant, therapy is critical to prevent fetal loss; aspirin (160 mg) is combined with prophylactic doses of either subcutaneous UFH (10,000 to 15,000 U per day in divided doses) or LMWH (to achieve an anti-Xa level of 0.1 to 0.3 U/mL). When such women have a history of TE disease, therapeutic doses of LMWH or UFH plus aspirin are employed. Heparin should be discontinued at the time of labor and delivery, although the risk for hemorrhage is not high during delivery, especially if anti-Xa levels are less than 0.7U/mL. One concern with residual anticoagulation at delivery is the risk for spinal hematoma with epidural anesthesia; this concern has been reported with both UFH and LMWH. The anti-Xa level that is safe for an epidural procedure is not known. Protamine sulfate can be used to neutralize UFH if the PTT is prolonged during labor and delivery; however, LMWH is only partially (10%) reversed by protamine.

Anticoagulation during the postpartum period can be carried out with heparin or warfarin; neither drug is contraindicated during breastfeeding. Women receiving long-term warfarin therapy (e.g., for valvular heart disease) who wish to become pregnant need to be switched to a fully anticoagulating dose of UFH or LMWH; warfarin treatment can be restarted postpartum.

Perioperative Anticoagulation A common clinical problem is the management of anticoagulation in patients who require surgery. The principles of care in this situation reflect the need for adequate hemostasis during and immediately after surgical procedures and the critical importance of restarting anticoagulation as soon as possible postoperatively, especially because surgery itself represents a relative hypercoagulable state. In patients with VTE who are anticoagulated on a short-term basis (3cm, ≤7cm, and located more than 2cm from the carina. No lymph nodes are involved. Tumor size is ≤7cm and located more than 2cm from carina. Peribronchial and/or hilar nodes may be involved. Tumor size is ≤7cm located more than 2cm from carina with peribronchial and/or hilar nodes involved; or tumor size >7cm (or satellite tumors within the same lobe) with invasion of chest wall, diaphragm, mediastinal pleura, pericardium, or located less than 2cm from the carina, without lymph node involvement. Any size tumor is present and may have invaded chest wall, carina, heart, great vessels, trachea, and esophagus. Tumor may involve ipsilateral peribronchial, hilar, mediastinal, and/or subcarinal nodes. Any size tumor is present and may have invaded any structure. Nodal involvement is always present and may extend to contralateral mediastinum or supraclavicular or scalene area. Metastases are present (including a malignant pleural effusion).

Small Cell Lung Cancer Limited

Extensive

Tumor is confined to one lung. Nodes may involve contralateral lung, but all cancer must be encompassed in one radiation portal. Metastatic disease is present or disease is not limited to one radiation field.

Non–Small Cell Lung Cancer The goals of staging for NSCLC are to find patients who may be cured by resection. Therefore, attention to the mediastinum, a common site of lymph node spread, and a search for metastatic disease are both performed as soon as possible after diagnosis. Testing usually begins with a CT scan extending through the liver and adrenal glands, common sites of metastases. Bronchoscopy or fine-needle aspiration is frequently used to make the histologic diagnosis. Patients with NSCLC who have enlarged mediastinal lymph nodes should undergo mediastinoscopy or bronchoscopic transbronchial biopsy to determine resectability (mediastinal nodes are rarely resectable). If an adrenal gland is enlarged, biopsy of the adrenal gland should be performed. In patients with resectable disease, surgical treatment may offer a chance for cure. Positron-emission tomography (PET) is a useful test during the work-up of the patient with NSCLC, but positive findings require pathologic or more precise radiologic corroboration before deciding against lung cancer surgery.

Small Cell Lung Cancer The goal of staging in SCLC is to determine which patients have limited-stage disease (about 30% to 40% of SCLC patients at the time of diagnosis) who can be potentially cured by the administration of combined chemoradiotherapy, compared with patients with extensive-stage metastatic

disease who cannot be cured but who can enjoy significant palliation with extension of survival as a result of chemotherapy. Limited-stage SCLC is defined as regional intra thoracic disease that can be encompassed within a single radiation field. Conversely, extensive disease represents locally advanced or widely metastatic disease that extends beyond a reasonable radiation field. Because common sites of metastatic disease include brain, liver, bone, and adrenal glands, diagnostic tests are performed to target these areas. The staging system for each type of lung cancer is shown in Table 57-2. For the patient with NSCLC, tumor size, proximity to central structures, and location of lymph nodes are the most important features. Because surgery is rarely performed for the patient with SCLC, nodal detail is seldom obtained. The regional extent of the cancer and the ability to encompass the involved areas within a single radiation field determine treatment planning (Table 57-3).

TREATMENT Non–Small Cell Lung Cancer Because complete removal of the tumor provides the best chance for long-term survival, the focus of the primary treat-

Chapter 57—Solid Tumors

605

Table 57-3 Treatment and Outcomes for Lung Cancer Tumor

Standard Treatment by Stage

Outcome

Non–small cell lung cancer

Early stages: surgery only (I); surgery followed by adjuvant chemotherapy (II, IIIA)

Small cell lung cancer

Later stage (unresectable): Combined chemotherapy and radiation or chemotherapy alone Limited: cisplatin-based chemotherapy with concurrent chest radiation Prophylactic cranial radiation is considered Extensive: chemotherapy palliates symptoms and prolongs survival but is not curative

Early stages: patients with stage II lung cancer have a 40%-50% survival rate with surgery and adjuvant chemotherapy Late stage: patients with stage IV lung cancer have 1-yr survival rate near 20% with chemotherapy Limited: 20%-30% 5-yr survival rate

ment for the patient with NSCLC should begin with an assessment of resectability. Resectability depends on the anatomic location of the tumor as well as the patient’s medical condition and pulmonary reserve. In general terms, the risk for pneumonectomy is small if the patient has a forced expiratory volume in 1 second greater than 2 L, carbon dioxide diffusing capacity of more than 60%, a maximal voluntary ventilation of more than 50% of predicted values, or the ability to walk up three flights of stairs. Lesser resections (e.g., lobectomy) may require less stringent criteria. Occasionally, patients with severe obstructive disease cannot undergo a curative procedure because they lack pulmonary reserve. Patients with stage I and II tumors (localized lesions or involvement limited to hilar lymph nodes) should undergo surgical treatment. Peripheral tumors are removed by lobectomy; more central tumors, if resectable, may require pneumonectomy. Stage III tumors may be operable in some patients, particularly if they are treated with chemotherapy and radiation therapy (i.e., neoadjuvant therapy) before resection. In patients with significant mediastinal lymph node involvement detected during tumor resection or at the time of mediastinoscopy, the chance of longterm survival is less than 20% even with surgical treatment. Although postoperative radiation therapy is seldom useful, recent reports show a survival benefit when patients with stages II and IIIA undergo adjuvant chemotherapy. Chemotherapy using a variety of agents combined with cisplatin or carboplatin improves survival after surgery by about 5%. Almost 80% of lung cancers are unresectable. If the tumor cannot be resected and has not spread to distant organs (stages IIIA and IIIB), chemotherapy is administered concurrently with radiation. This treatment leads to better median survival and 5-year disease-free survival rates than with radiation alone, but combined therapy for unresectable disease should be reserved for patients with good functional status. Aggressive treatment is much less effective (due to unacceptable toxicity) in patients who have lost more than 5% of their body weight or who are active less than 50% of the day. The median survival for patients with locally advanced, unresectable lung cancer who are candidates for combined modality therapy is now 15 to 18 months, and nearly 20% achieve a 5-year disease-free survival. Patients with metastatic disease may benefit from chemotherapy. The most active agents are platinum derivatives

Extensive: median survival rate is 10mo with treatment

such as cisplatin and carboplatin, taxanes such as paclitaxel and docetaxel, gemcitabine, vinorelbine, irinotecan, and pemetrexed. About 20% to 30% of patients achieve a greater than 50% reduction in their tumor volume after chemotherapy. For patients with advanced-stage lung cancer (stages IIIB with pleural effusion and IV), chemotherapy improves survival from an average of 8.5 to 11 months; quality-of-life studies show that chemotherapy delays symptoms and reduces their severity compared with no treatment. New agents that target neoplastic angiogenesis such as bevacizumab or the epidermal growth factor receptor antagonists such as erlotinib and cetuximab have shown therapeutic benefit either as single agents (e.g., erlotinib) or, more commonly, in combination with chemotherapy (bevacizumab) for these cancers.

Small Cell Lung Cancer Patients with limited-stage SCLC are treated with curative intent with combined (concurrent) modality chemotherapy and radiotherapy. It is common practice to treat these patients with four to six cycles of chemotherapy (most commonly with cisplatin or carboplatin and etoposide) along with radiation therapy delivered to a radiation field encompassing the intrathoracic disease. About 50% to 60% of patients experience a complete response, and another 20% to 30% have a partial shrinkage of their cancer. Because 50% to 60% of patients with SCLC will ultimately develop brain metastases, prophylactic cranial radiation should be considered for those patients who have a complete response in the lung to primary chemoradiotherapy. About 20% to 30% of treated patients with limited-stage SCLC are alive and free of disease 3 years after diagnosis, and a proportion of these individuals are cured. Although patients with extensivestage SCLC cannot be cured with existing therapy, chemotherapy can cause rapid tumor shrinkage with resolution of symptoms resulting in significant prolongation of life compared with supportive care alone.

Recurrent Lung Cancer Both NSCLC and SCLC have high relapse rates. The use of second-line chemotherapy for these patients may palliate symptoms, but its use is not curative. Although these cancers may respond to subsequent agents such as docetaxel, pemetrexed, and erlotinib (NSCLC) or the camptothecins, the taxanes, and oral etoposide (SCLC), the use of these drugs should be limited to patients with good performance status, preferably on a clinical trial.

606

Section X—Oncologic Disease

Head and Neck Cancers EPIDEMIOLOGY AND NATURAL HISTORY Most cancers of the head and neck, including cancers of the larynx, oral cavity, oropharynx, and sinuses, are squamous cell carcinomas. In 2008, an estimated 35,310 new cases were diagnosed, and 7590 deaths occurred. Tobacco use, alcohol consumption, and poor oral hygiene have all been linked to the development of cancers of the head and neck. Nasopharyngeal cancers are associated with Epstein-Barr viral infection. Infection with human papillomavirus (HPV), in particular HPV-16, is now an established risk factor for oropharyngeal cancer. The major determinant of prognosis is the tumor burden, or the thickness of the tumor, and the presence or absence of regional lymph node involvement. The cure rate with small tumors is as high as 75% to 95% with radiation therapy or surgery. Continued use of tobacco after a diagnosis of a head and neck cancer is associated with a poor prognosis. People who have had cancer of the head and neck are at high risk for having a primary lung or esophageal cancer.

SYMPTOMS Symptoms of cancers of the head and neck are related to the location of the tumor. For example, supraglottic laryngeal cancers cause pain with swallowing and a change in voice quality. Cancers of the oral cavity may exhibit a mass under the tongue or red or white patches in the mouth. Bleeding in the mouth or ill-fitting dentures may also be symptoms of cancer in the oral cavity. Symptoms of cancers of the sinus include sinusitis that does not resolve with appropriate treatment. Ear pain may be present with cancers of the oropharynx or hypopharynx.

DIAGNOSIS

may also derive palliative benefit from combined modality chemotherapy (or cetuximab) and radiotherapy.

Gastrointestinal Cancers Cancers of the gastrointestinal tract are among the most common tumors, with more than 271,000 new cases occurring in 2008. Advances in the treatment of colorectal cancer have improved survival and quality of life for patients with these diseases. Table 57-4 outlines the common signs and symptoms, treatments, and prognosis of gastrointestinal tumors.

ESOPHAGEAL CANCER Epidemiology and Natural History The two types of esophageal cancer are squamous cell and adenocarcinoma. Squamous cell cancers are most common in the cervical and thoracic esophagus, and adenocarcinomas commonly occur in the lower esophagus down to the gastroesophageal junction. Squamous cell cancers are more common in African Americans and are associated with predisposing factors that include smoking, caustic injury, achalasia, and alcohol intake. Squamous cell cancers are associated with other tobacco-related cancers in the upper airways and digestive tract. The rate of adenocarcinoma is increasing; this increase is related in part to Barrett esophagus, an adenomatous metaplasia of the distal esophagus often caused by gastroesophageal reflux disease, but other factors are likely. Almost 25% of patients with severe Barrett esophagus eventually develop esophageal adenocarcinoma. The most useful intervention for patients with Barrett esophagus is frequent endoscopic screening and biopsy; pharmacologic treatment of acid reflux disease does not prevent neoplastic transformation.

Diagnosis of head and neck cancers require histologic confirmation with biopsy. Magnetic resonance imaging (MRI) or a CT scan of the head and neck is performed to determine a precise estimate of the extent of the tumor. Thorough examination of the entire aerodigestive tract with endoscopy will demonstrate a synchronous second primary tumor, for example, in the esophagus, in up to 15% of patients. Staging of the tumor is based on clinical and radiographic assessment.

Symptoms

TREATMENT

An upper gastrointestinal radiographic series or endoscopy will demonstrate an esophageal lesion, which should then undergo biopsy. The most effective staging tool is endoscopic ultrasonography, an accurate tool in assessing local depth of penetration and lymph node metastases. CT and positron-emission tomographic (PET) scanning are also required to ensure that the tumor has not already metastasized to the chest or liver, the two most common sites of spread.

Small tumors that have not spread to regional lymph nodes are treated with radiation or surgery. Primary radiation therapy may allow preservation of organ function of, for example, the larynx, with surgical resection used in the case of recurrence. Locally advanced disease is treated with a combination of surgery and radiation therapy (with or without chemotherapy), or with a combination of radiation therapy and cisplatin-containing chemotherapy regimens. Most relapses occur within 2 to 3 years after therapy. Close surveillance is therefore warranted. Good-performance status patients with locally advanced, unresectable disease

The most common symptom of esophageal cancer is dysphagia. As the lumen of the esophagus narrows, the patient loses normal swallowing capacity and has a sensation that solid food becomes “stuck.” Eventually, the patient may be unable to swallow liquids. Patients commonly become afraid to eat because of frequent regurgitation at mealtime, resulting in significant weight loss.

Diagnosis

Treatment The most common treatment of esophageal cancer is surgery. Resection of the involved esophagus includes a wide margin

Chapter 57—Solid Tumors

607

Table 57-4 Gastrointestinal Cancers Tumor Site

Common Findings

Standard Treatments

Expected Outcome

Esophageal

Dysphagia, chest pain, weight loss

Stage II/III stage: about 30% 5-yr survival Average survival for metastatic disease is 90% 5-yr survival Positive nodal involvement: 20%-75% 5-yr survival for tumors 70% at 5yr Nodal involvement: 50%-70% at 5yr Metastatic: median 14-24mo

Localized: 70% at 5yr

posing conditions include pernicious anemia, achlorhydria, gastric ulcers, and prior gastric surgery. Except for cancers of the gastroesophageal junction, gastric cancer rates have decreased in the United States (21,500 new cases occurring in 2008). A recognized risk factor for gastric cancer is infection with Helicobacter pylori. Whether early treatment of H. pylori infection changes the rate of cancer in infected populations is not clear.

Diagnosis Patients with gastric cancer commonly experience abdominal pain, early satiety, anemia, hematemesis, weakness, and weight loss. Frequently, the cancer has already involved local lymph nodes by the time the diagnosis is made. Physical examination may show a gastric mass, an umbilical node (Sister Mary Joseph node), or a left supraclavicular node (Virchow node). Pathologic analysis shows an adenocarcinoma that can be localized or spread throughout the gastric lining (linitis plastica). Required staging includes a CT scan to search for obvious nodal or metastatic involvement of the liver, upper gastrointestinal endoscopic examination, and endoscopic ultrasonography to determine depth of invasion and biopsy abnormal lymph nodes.

Treatment Gastric cancer is most often treated surgically. When the tumor and all relevant lymph nodes have been removed, patients have a 20% to 60% chance of a 5-year survival, depending on the pathologic stage. If gastric cancer recurs,

608

Section X—Oncologic Disease

the most common sites are local extension or hematogenous spread through the portal vein to the liver. Patients undergoing a complete resection for gastric cancer benefit from the addition of 5-fluorouracil (5-FU) and leucovorin chemotherapy and postoperative radiation therapy. This combination improves median survival by about 15 months compared with no adjuvant therapy. Patients with metastatic gastric cancer may elect chemotherapy to palliate symptoms. Active agents include platinum compounds, fluoropyrimidines, anthracyclines, taxanes, and irinotecan. Combination chemotherapy provides a 20% to 40% response rate and may extend survival.

COLORECTAL CANCER Epidemiology and Natural History About 1 in 20 people in the United States will be diagnosed with colon cancer (lifetime risk is 6%). An estimated 148,700 new cases were diagnosed in 2008, and in the same year, nearly 50,000 deaths occurred. Known predisposing factors are a history of ulcerative colitis and a strong family history of colon cancer. Several mutations, whether inherited or spontaneous, play a major role in the predisposition to colon cancer (see Chapter 55, Fig. 55-1 and Table 55-2). For example, familial polyposis is transmitted in an autosomal dominant manner. Individuals have mutations in the APC gene, which may be associated with periampullary and thyroid cancers or non-neoplastic growth such as osteomas, sebaceous cysts, and gastric polyps. Hereditary nonpolyposis colorectal cancer (HNPCC) is a more common autosomal disorder associated with microsatellite instability and mutations in hMSH-2, hMLH-1, PMS-1, PMS-2, and hMSH-6. Patients with HNPCC usually have colon or endometrial cancer when younger than 50 years and have first-degree relatives with colon cancer or other HNPCC-related cancers derived from the stomach, ovary, small bowel, biliary tract, ureter, or renal pelvis. Whether the predisposing risk for colon cancer is genetically transmitted or sporadically acquired, a clear relationship exists between adenomatous polyps and the later development of colon cancer. Because the removal of polyps is probably the most important way to prevent the development of invasive colon cancer, the most reliable way to reduce colon and rectal cancer mortality is to perform regular screening. Colonoscopy is the most commonly used screening test. Studies using sigmoidoscopy and regular fecal occult blood testing also show reductions in the incidence and mortality of colorectal cancer. Patients with a proven mutation (Gardner syndrome, HNPCC) or a strong family history (familial adenomatous polyposis), and those who acquire other diseases associated with colorectal cancer, such as ulcerative colitis, should begin colonoscopy earlier than suggested for the general population. For patients with familial adenomatous polyposis, screening should start in the teenage years. For patients with HNPCC, screening for colon cancer should start 10 years before the age of diagnosis in the youngest family member with colorectal cancer. Research efforts are underway in the primary prevention of colorectal cancers using interventions such as diet, daily aspirin, cyclo-oxygenase-2 inhibitors, calcium and vitamin D supplementation, and other chemopreventive agents to reduce cancer incidence. Enthusiasm for promoting a high-

fiber diet to reduce the risk for colon cancer has waned. Lifestyle changes are still considered important—fresh fruits and vegetables, regular exercise, fewer than two red meat servings per week—based on epidemiologic association.

Symptoms Rectal bleeding commonly occurs with colon and rectal cancers. Patients with left-sided colon lesions often complain of a change in stool color or caliber or pelvic pain and transient bloating. Right-sided lesions may become friable and may result in occult bleeding. Occasionally, patients with colon and rectal cancers are asymptomatic until the tumor totally obstructs the bowel or perforates the peritoneal cavity. Colon and rectal cancers tend to spread hematogenously to the lungs and liver. Rectal cancer is more likely than colon cancer to recur locally because it is more difficult to get a wide margin of normal tissue and lymph nodes within the tight confines of the pelvis.

Diagnosis The work-up for colon cancer requires measurement of serum carcinoembryonic antigen (CEA), an abdominalpelvic CT scan, a chest radiograph, and endoscopic imaging of the colon to ensure that all polyps and cancers are removed near the time of the primary operation. Table 57-5 describes the staging system for colon and rectal cancers.

Treatment In patients with stages I, II, and III colon cancer, surgical resection of the primary carcinoma along with any regional lymph node metastases is routinely performed; multiagent, adjuvant chemotherapy is recommended for all patients with stage III cancer (which reduces the rate of recurrence by about 40%) and for selected high-risk individuals with stage II cancer. For patients with rectal cancer, either primary resection (which may require colostomy) or neoadjuvant chemoradiotherapy is performed. Because of the higher likelihood of local recurrence in rectal cancer, any lesion that invades the muscle or lymph nodes is also treated with radiation therapy (if not given before surgery) and adjuvant chemotherapy. Surgical resection of the primary lesion in patients with advanced (stage IV) colorectal cancer

Table 57-5 Staging for Colon and Rectal Cancer Stage

Tumor Size

Nodal Status

Metastases

0 I

In situ Invades mucosa only May invade muscularis or through serosa Any size tumor or any level of invasion Any size or depth

No No

Yes

Positive nodal involvement present or absent

Yes

III IV

Chapter 57—Solid Tumors may be recommended to palliate or avoid symptoms of obstruction, bleeding, and pain. Significant advances have been made in the use of chemotherapy for colorectal cancer. Regimens proved successful in prolonging survival for colon cancer include 5-FU with leucovorin or its oral analogue, capecitabine, alone. The addition of oxiliplatin to 5-FU–based adjuvant therapy has recently been shown to further improve survival after surgery. In the setting of metastatic cancer, infusions of chemotherapy with FOLFOX (5-FU, leucovorin, oxiliplatin), or FOLFIRI (5-FU, leucovorin, irinotecan) with the addition of antibodies targeting the vascular endothelial growth factor (VEGF; bevacizumab) or the epidermal growth factor receptor (cetuximab) prolong median survival beyond 20 months, almost twice the survival expected in the early 1990s. Despite improvements in survival, metastatic colon and rectal cancers are incurable unless the metastatic lesions can be surgically resected.

ANAL CARCINOMA Anal cancers are increasing in frequency (currently about 5000 new cases per year). Persons infected with HPV and HIV are more likely to develop anal cancer. Patients with anal cancer usually experience rectal bleeding or complain of rectal fullness. Combined chemotherapy with 5-FU and mitomycin and radiation therapy make up the standard approach to a patient with localized anal cancer. Results with combined therapy are superior to those with surgery, with the additional benefit of sparing the anal sphincter. Abdominoperineal resection is reserved for patients in whom chemoradiotherapy fails.

PANCREATIC CANCER Epidemiology and Natural History Pancreatic cancer, which is diagnosed in more than 37,000 people living in the U.S. each year, is strongly associated with cigarette smoking. A small proportion of pancreatic cancers are inherited from mutations in the p16 and BRCA-2 genes. Epithelial pancreatic cancer is an adenocarcinoma with an extremely high mortality rate because it is usually diagnosed when the tumor is beyond the capability of surgical resection. A less common type of pancreatic cancer, islet cell carcinoma, originates in the endocrine cells. Symptoms related to secretion of peptides such as gastrin, vasointestinal polypeptide, and insulin characterize these tumors.

Symptoms The most common presentation of epithelial pancreatic cancer is abdominal pain accompanied by rapid weight loss. Characteristically, the pain is located in the periumbilical region and pierces or stabs through to the back. The pain is often explained by the frequent invasion of the celiac plexus deep in the retroperitoneum. Other symptoms of pancreatic cancer are the recent onset of diabetes, intestinal angina reflecting encasement of the superior mesenteric artery, a palpable gallbladder (Courvoisier sign), and jaundice from blockage of the distal common bile duct. Migrating thrombophlebitis (Trousseau sign) is a paraneoplastic complication of pancreatic adenocarcinoma. The tumor

609

marker CA-19-9 is elevated in up to 75% of patients with pancreatic cancer.

Treatment The only curative treatment for pancreatic cancer is pancreaticoduodenectomy (Whipple procedure), an extensive operation requiring three anastomoses that carries a high mortality rate in centers with less experience with the procedure. The 5-year survival for surgically treated patients with localized pancreatic cancer is 25% for node-negative cancer but only 10% when lymph nodes are involved. Accordingly, it has become standard practice to offer adjuvant chemoradiotherapy or gemcitabine chemotherapy alone, which may provide a survival advantage. Patients with unresectable disease may benefit from local radiation therapy with concurrent 5-FU; more than 30% of patients treated with this combination have some improvement in their symptoms. Alternatively, gemcitabine-based chemotherapy alone may have palliative benefit. When patients have progressive or metastatic disease, the use of palliative chemotherapy with weekly gemcitabine has been shown to improve quality of life and survival to a small degree (5.7 months with gemcitabine, 4.4 months without gemcitabine, or 20% 1-year survival versus 5%). Recent meta-analyses suggest that certain gemcitabine-based chemotherapy combinations offer limited survival advantage over gemcitabine alone.

HEPATOCELLULAR CARCINOMA Although uncommon in the United States (about 21,000 new cases per year), hepatocellular carcinoma (HCC) is one of the most common cancers throughout the world; more than 1 million cases are diagnosed each year. The common causes of HCC are chronic viral hepatitis (both B and C) and cirrhosis related to alcohol use or hemochromatosis. Although this approach is unproved, considerable interest exists in screening patients who are at extremely high risk with serial measurement for α-fetoprotein (AFP) levels. AFP levels are commonly elevated even in early-stage HCC. Treatment of early-stage HCC is surgery. Cure rates are more than 75% for patients with tumors smaller than 2 cm. Patients with severe cirrhosis and who have small liver cancers may benefit from liver transplantation. Chemoembolization may provide palliative benefit for patients with unresectable tumors, but conventional cytotoxic chemotherapy is generally ineffective. However, recent findings suggest that molecularly targeted tyrosine kinase inhibitors such as sorafenib offer a survival advantage over supportive care.

Breast Cancer EPIDEMIOLOGY Breast cancer is the most common nonskin cancer in women and the second leading cause of cancer death (after lung cancer) among women in the United States. In 2008, an estimated 182,460 women were diagnosed with invasive breast cancer, and more than 40,000 died from breast cancer. Breast cancer in men is rare. Risk factors for breast cancer include older age, a family history of breast cancer, early menarche, late menopause,

610

Section X—Oncologic Disease

first-term pregnancy after age 25 years, nulliparity, prolonged use of exogenous estrogen, and postmenopausal obesity. Exposure to ionizing radiation, as is used in the treatment of Hodgkin disease, also increases the risk for breast cancer. Only 5% to 10% of patients with breast cancer are associated with the breast cancer-susceptibility genes, BRCA-1 and BRCA-2. Patients with multiple affected family members or those with a personal or family history of male breast cancer, bilateral breast cancer, or ovarian cancer should be offered genetic counseling and genetic testing for BRCA-1 and BRCA-2. Mammographic screening in both average- and high-risk populations has been shown to reduce breast cancer mortality.

PATHOLOGY Most breast cancers are infiltrating ductal adenocarcinomas. A smaller proportion of breast cancers are infiltrating lobular adenocarcinomas. Tubular and mucinous carcinomas, which are a subtype of infiltrating ductal cancers, are associated with a better prognosis. The estrogen and progesterone receptor status of the primary tumor should be assessed in all cases of invasive breast cancer. The oncoproteinoncogene HER-2/neu (human epidermal growth factor receptor) is important in defining prognosis and treatment. Tumors that are negative for the estrogen and progesterone receptor and for overamplification of the HER-2 oncogene (i.e., the so-called triple-negative tumors) are associated with a poorer prognosis. Such tumors are characteristic of those that develop in women who have the BRCA-1 breast cancer susceptibility gene. Ductal carcinoma in situ (DCIS), also called intraductal carcinoma, is increasing in frequency, most likely because of increased mammographic screening.

CLINICAL PRESENTATION Breast cancer is most often diagnosed through screening mammography or after a patient or her physician notices a palpable mass. Fewer than 10% of women have metastatic disease at diagnosis. Recurrent breast cancer most commonly exhibits metastases in the bone, liver, lung, and central nervous system, but breast cancer can recur in any organ of the body. Women with a history of breast cancer are also at increased risk for breast cancer in the contralateral breast. Inflammatory breast cancer is a clinical diagnosis in a woman with breast induration and erythema, often without a palpable mass. The skin findings are due to tumor emboli in the dermal lymphatics; skin biopsy is negative for cancer in 50% of patients.

STAGING Breast cancer staging requires removal of the primary tumor and ipsilateral axillary lymph node assessment. Women with tumors larger than 5 cm and those with positive axillary lymph node involvement may have additional radiographic tests, including a chest radiograph, bone scan, and CT scan of the abdomen. Patients with smaller tumors and negative lymph node involvement do not need these tests unless they exhibit symptoms (e.g., skeletal pain) suggestive of metastatic involvement.

TREATMENT For women with small breast tumors, breast-conserving therapy with lumpectomy followed by radiation therapy is standard therapy. Women with large tumors or with two or more tumors in separate quadrants of the breast should undergo mastectomy. Some women who are candidates for breast-conserving therapy choose to have a mastectomy, with or without breast reconstruction. Women who have had previous radiation to the breast, either for a previous breast cancer or other malignancies, are generally treated with mastectomy. Chemotherapy administered before surgical treatment (primary or neoadjuvant chemotherapy) may allow breast conservation in women with large tumors who would otherwise not be able to undergo a lumpectomy. Preoperative hormone therapy can be considered in frail patients with hormone receptor–positive tumors, but hormone therapy does not replace surgical treatment for most patients. Adjuvant therapy with chemotherapy and hormonal therapy improves relapse-free and overall survival rates in premenopausal and postmenopausal women who are at high risk for metastatic relapse. The monoclonal antibody trastuzumab, directed at the HER-2 pathway, improves disease-free survival in patients with tumors that have overamplification (or high levels of overexpression) of the HER-2 oncoprotein. Treatment decisions in women with metastatic disease are based on the hormone receptor status, sites of disease, presence and severity of symptoms, time since initial diagnosis, and previous treatments. The monoclonal antibody trastuzumab may be used in combination with chemotherapy, hormonal therapy, or as a single agent. Life expectancy is longer in women with hormone-responsive disease and with lymph node or bone metastases, rather than liver, lung, or central nervous system metastases. Patients with metastatic breast cancer may live for many years, often responding to hormonal therapy for years before requiring chemotherapy for disease control. Many chemotherapeutic agents (singly or in combination), including the anthracyclines, taxanes, alkylating agents, fluoropyrimidines, vinca alkaloids, gemcitabine, and epithilones, demonstrate activity against breast cancer. Bisphosphonates, such as zoledronate and pamidronate, are given intravenously to decrease the pain associated with bone metastases and the risk for fracture in women with skeletal metastases. DCIS is treated with either lumpectomy followed by radiation therapy or mastectomy. Women with multifocal or palpable DCIS should have assessment of lymph node status because a small but measurable proportion will have positive lymph node involvement that suggests the presence of invasive cancer. In these patients, systemic treatment is identical to that given to women with a clearly invasive breast primary tumor. Prophylactic bilateral mastectomy is offered to women with the BRCA-1 or BRCA-2 breast cancer susceptibility genes. An alternative approach is close clinical surveillance, including monthly self-examination, frequent examination by a physician, and regular mammography or MRI. Oophorectomy or antiestrogen therapy can be used to decrease the risk for breast cancer in these women and in other women at high risk for the disease (see Chapter 56).

Chapter 57—Solid Tumors

611

Table 57-6 Genitourinary Cancers Tumor Site Testicle

Prostate

Bladder

Renal cell

Common Findings

Standard Treatments

Expected Outcomes

Testicular swelling, pain, back pain Cough with metastatic presentation Elevated prostatespecific antigen, decreased urinary stream Bone pain with metastatic presentation Hematuria, cystitis

Inguinal (not scrotal) orchiectomy Node-positive seminoma: radiation therapy Node-positive NSGCT: RPLND or chemotherapy

Early-stage seminoma: >90% 5-yr survival Stage III NSGCT: about 75% 5-yr survival Poor-risk tumors: 35 inches in women), glucose intolerance (fasting plasma glucose ≥100 mg/dL), hypertension (≥130/≥85 mm Hg), elevated plasma triglycerides (≥150 mg/dL), and low high-density lipoprotein (HDL) cholesterol (0.5g/day or >3+ if quantitation is not performed. OR b. Cellular casts: May be red cell, hemoglobin, granular, tubular, or mixed. a. Seizures: Occur in the absence of offending drugs or known metabolic derangements (e.g., uremia, ketoacidosis, electrolyte imbalance) OR b. Psychosis: Occurs in the absence of offending drugs or known metabolic derangements (e.g., uremia, ketoacidosis, electrolyte imbalance) a. Hemolytic anemia: Develops with reticulocytosis. b. Leukopenia: 6.5mg/kg, and obesity. Patients should have baseline and annual ophthalmologic examinations with visual field testing. Azathioprine and methotrexate are both immunosuppressive agents prescribed in patients with lupus either when glucocorticoids alone are not fully effective or to allow for a reduction in the glucocorticoid dose. Toxicities of azathioprine include cytopenias, increased infection risk, and potential association of hematologic malignant disease (controversial). Azathioprine may be used during pregnancy for severe internal organ lupus (especially nephritis). Methotrexate is particularly effective with inflammatory arthritis associated with lupus. In addition to cytopenias and infections, liver function abnormalities, alopecia, nausea, and pneumonitis are potentially seen with methotrexate. Because it is teratogenic, methotrexate should be stopped 3 to 6 months before pregnancy. Both drugs require monthly laboratory monitoring. Mycophenolate mofetil (MMF) is increasingly used to treat patients with internal organ involvement of lupus, particularly nephritis. MMF inhibits purine synthesis of lymphocytes and has been standard treatment in the prevention of solid-organ transplant rejection. Recent clinical trials have shown that MMF is as efficacious as intravenous (IV) cyclophosphamide (described later) for inducing remission of active lupus nephritis. Toxicity of MMF includes gastroin-

testinal disturbance and leukopenia; it also has a category D rating in pregnancy. Patients with neurologic lupus, rapidly progressive nephritis, or vasculitis of internal organ systems are often treated with cyclophosphamide, the most potent immunosuppressive agent (alkylating agent) used to treat SLE. Because it is extremely toxic, this drug is generally reserved for the most severe disease manifestations of lupus. In a series of studies, primarily conducted at the National Institutes of Health, monthly IV cyclophosphamide was effective in reducing the rate of progression of lupus nephritis to endstage renal disease. Acute toxicities of cyclophosphamide include pancytopenia, alopecia, mucositis, and hemorrhagic cystitis. Long-term cyclophosphamide use may lead to transitional cell carcinoma, hematologic malignant disease, sterility, premature menopause, and opportunistic infections. Many other treatments have been used in SLE, although extensive clinical trials demonstrating efficacy are lacking. Among these drugs, intravenous immunoglobulin may be effective for SLE-related thrombocytopenia and catastrophic antiphospholipid syndrome. Thrombotic thrombocytopenic purpura secondary to SLE may be treated with plasmapheresis. Immunoablation with high-dose cyclophosphamide with or without autologous stem cell transplant has been tried in a small number of patients with progressive active disease that is refractory to conventional immunosuppressive therapy; unfortunately mortality is high, and data are preliminary. Great potential and optimism exist for “biologic” immunomodulating agents that focus on various aspects of the immune system, including B cells, interactions between B and T cells, and cytokines. The most promising current therapeutic agents are those that deplete B cells, producers of autoantibodies. Among these, rituximab has been successfully used in refractory lupus patients with hematologic and renal disease manifestations.

Special Issues in the Care of Patients with Systemic Lupus Erythematosus PREGNANCY Pregnant women with lupus have higher rates of both pregnancy loss (miscarriage and stillbirth) and preterm delivery (premature rupture of membranes, preeclampsia, and intrauterine growth restriction). Lupus activity preceding conception, especially nephritis, hypertension, and antiphospholipid antibody syndrome, are clear risk factors for pregnancy complications in SLE. Pregnancy itself may place women with lupus at greater risk for a flare, particularly if the disease, specifically nephritis, was active before conception. Neonatal lupus is a rare disorder in which maternal antiSSA/Ro-SSB/La antibodies cross the placenta and injure the fetus. Mothers with these autoantibodies have a 2% risk of having a child with congenital heart block (CHB). These mothers are screened with fetal heart tones and fetal echocardiography between weeks 16 and 34 of gestation. Treatment with a fluorinated corticosteroid (dexamethasone)

Chapter 81—Systemic Lupus Erythematosus may be beneficial; however, a considerable number of children with CHB do not survive (30%) or experience morbidity, with over 60% requiring pacemakers. More common manifestations of neonatal lupus are skin rashes, cytopenias, and hepatosplenomegaly, all of which resolve within 6 to 8 months (after maternal autoantibodies are removed from the child’s circulation). Mothers of children with neonatal lupus do not necessarily have SLE or Sjögren syndrome; however, some of these women eventually are diagnosed with an autoimmune disease. With careful prenatal screening and planning, women with lupus can successfully have a healthy child. Prenatal monitoring of anti-SSA/Ro-SSB/La antibodies and antiphospholipid and anticardiolipin antibodies and a prepregnancy high-risk obstetric consultation are critical. Ideally, women with lupus should have clinical quiescence for 6 months before considering a planned pregnancy. Special consideration must also be given to medications, as the majority of immunosuppressants (methotrexate, mycophenolate mofetil, and cyclophosphamide) are teratogenic.

HORMONAL THERAPY Hormones are thought to play a role in the development of SLE, given its female predominance. Consequently, rheumatologists have historically been hesitant to prescribe estrogen therapy for fear of inducing a flare. Recently, randomized, placebo-controlled clinical trials, described in the next paragraphs, have helped to guide hormonal therapy in women with lupus. A multicenter randomized trial revealed that oral contraceptive therapy does not appear to increase the risk of SLE flares in women with mild or stable SLE disease activity. However, this is not generalizable to all women with SLE, in particular those whose disease is active or severe and those with prior thrombotic events or antiphospholipid and anticardiolipin antibodies. Clearly, an effective form of birth control is necessary for young sexually active women with SLE, especially those taking teratogenic medications. Physicians must carefully discuss the risks and benefits of birth control with lupus patients. Hormone replacement therapy (HRT) is a controversial topic, regardless of lupus status. However, it is of particular interest in SLE because some women with lupus reach menopause prematurely. In a recent clinical trial of HRT in SLE patients with mild or stable disease, with no prior thrombotic events, antiphospholipid and anticardiolipin antibodies, or malignancies of women, no lupus patients had severe clinical flares, but 20% did have mild to moderate flares. These findings suggest that brief (1 year) HRT may have a role in alleviating menopausal symptoms or in treating osteoporosis in a certain subset of women with SLE.

CARDIOVASCULAR DISEASE As survival and therapies for lupus have improved, CVD has emerged as a leading cause of morbidity and mortality. Lupus patients are 5 to 10 times more likely than healthy individuals to have a coronary event. More striking, premenopausal women aged 35 to 44 are more than 50 times as likely as healthy women to experience a myocardial infarc-

839

tion. Autopsy series have revealed atherosclerotic heart disease as the underlying mechanism of CVD in SLE. The cause of this premature atherosclerosis in lupus is multifactorial and includes inflammatory mediators, lupus-related factors (premature menopause, corticosteroid therapy, disease activity), and traditional cardiovascular risk factors. Prevention of CVD is a critical aspect of the care of lupus patients. Although no firm CV management guidelines exist for SLE patients, many argue that lupus patients should be considered and treated as coronary heart disease riskequivalents. Cardiac disease must be thought of and aggressively evaluated in SLE patients with both typical and atypical cardiac symptoms, regardless of age.

MALIGNANCY Recently, a multicenter international cohort of nearly 10,000 lupus patients reported an increased risk of malignancy in SLE compared with the general population. Most striking was a threefold to fourfold increased risk of non-Hodgkin lymphoma. Other hematologic, lung, and hepatobiliary cancers were also described with increased frequency in SLE patients. Malignancy risk appears to be highest early in the disease course, but malignancy is clearly a risk throughout a patient’s life span. Older age, tobacco use, SLE disease damage, and immunosuppressant use appear to be associated with cancers in SLE; however, the exact mechanism of malignancy in SLE is not yet fully elucidated. Although lymph node enlargement is a common manifestation in lupus patients, physicians must have a high index of suspicion for malignancy if the lymphadenopathy does not resolve with lupus treatment, is tender or nonmobile, or occurs without other lupus symptoms.

BONE HEALTH Lupus patients have higher rates of low bone mineral density (BMD), osteoporosis, and fractures, than do healthy agematched subjects. The increased risk is accounted for by both traditional risk factors, such as female sex, white race, older age, and low body weight, and lupus-associated factors. Fatigue and articular symptoms secondary to SLE may limit physical activity, leading to loss of bone strength. Furthermore, therapies commonly used in lupus contribute to overall loss of bone health. Corticosteroids reduce bone mass and are an independent risk factor for fractures in women with SLE. Cyclophosphamide use can lead to premature ovarian failure, another risk factor for osteoporosis. In addition, lupus disease damage, regardless of steroid use, leads to low BMD, suggesting that SLE itself is a risk factor for low BMD. Another contributing factor may be that lupus patients are commonly asked to avoid sun exposure, which can lead to low vitamin D levels and insufficient absorption of calcium. Given these SLE-specific risk factors for low BMD, prevention of osteoporosis is extremely important. Although osteoporosis screening guidelines for SLE patients are the same as for the general population, bone density scans should also be considered in patients with premature menopause and those who are or who will be on chronic (>3 months) corticosteroids.

840

Section XVI—Musculoskeletal and Connective Tissue Disease

Prospectus for the Future • Advances in the understanding of the immunopathologic features of SLE, providing a logical framework for specific targeting and testing of new biologic therapies. • Clinical trial data to evaluate the efficacy of new biologic agents for the treatment of SLE.

References Arbuckle MR, McClain MT, Rubertone MV, et al: Development of autoantibodies before the clinical onset of systemic lupus erythematosus. N Engl J Med 349:1526-1533, 2003. Austin HA 3rd, Klippel JH, Balow JE, et al: Therapy of lupus nephritis. Controlled trial of prednisone and cytotoxic drugs. N Engl J Med 314:614-619, 1986. Bernatsky S, Boivin JF, Joseph L, et al: An international cohort study of cancer in systemic lupus erythematosus. Arthritis Rheum 52:1481-1490, 2005. Buyon J, Petri, MA, Kim MY, et al: The effect of combined estrogen and progesterone hormone replacement therapy on disease activity in systemic lupus erythematosus: a randomized trial. Ann Intern Med 142:953-962, 2005. Domsic RT, Ramsey-Goldman R, Manzi S: Epidemiology and classification of SLE. In: Hochberg MC, Silman AJ, Smolen JS, et al (eds): Rheumatology, 4th ed. Philadelphia, Mosby, 2008.

• Development of formal management guidelines for osteoporosis and cardiovascular care in patients with SLE.

Ginzler EM, Dooley MA, Aranow C, et al: Mycophenolate mofetil or intravenous cyclophosphamide for lupus nephritis. N Engl J Med 353:2219-2228, 2005. Lee C, Almagor O, Dunlop DD, et al: Disease damage and low bone mineral density: An analysis of women with systemic lupus erythematosus ever and never receiving corticosteroids. Rheumatology (Oxford) 45:53-60, 2006. Lee C, Ramsey-Goldman R: Bone health and systemic lupus erythematosus. Curr Rheumatol Rep 7:482-489, 2005. Manzi S, Meilahn EN, Rairie JE, et al: Age-specific incidence rates of myocardial infarction and angina in women with systemic lupus erythematosus: comparison with the Framingham Study. Am J Epidemiol 145:408-415, 1997. Petri M, Kim MY, Kalunian KC, et al: Combined contraceptives in women with systemic lupus erythematosus. N Engl J Med 353:2550-2558, 2005.

Chapter

XVI

Antiphospholipid Antibody Syndrome

Surabhi Agarwal and Amy H. Kao

ntiphospholipid antibody syndrome (APS) is a disorder characterized by recurrent vascular thrombosis and/or recurrent pregnancy loss in the presence of positive antiphospholipid (aPL) antibodies. The aPL antibodies were initially identified in patients with falsepositive syphilis screening test results (positive result in the absence of syphilis). Lupus anticoagulant (LAC) was later described in patients with systemic lupus erythematosus (SLE), which was also frequently associated with a falsepositive result of a screening test for syphilis and was paradoxically associated with thrombosis and recurrent pregnancy loss. LAC is a misnomer because this in vitro anticoagulant effect reflects the prolonged activated partial thromboplastin time (aPTT), and the presence of LAC does not indicate a diagnosis of lupus. Furthermore, both LAC and anticardiolipin (aCL) antibodies were shown to bind to phospholipid-binding plasma proteins and not directly to phospholipids.

Pathogenesis Animal studies suggest that aPL may be directly pathogenic. The linear correlation between antibody levels and risk of thrombosis further supports their pathogenic role. aPL can cause β2-glycoprotein I (β2-GPI)–dependent endothelial activation by upregulation of expression of adhesion molecules and secretion of proinflammatory cytokines. β2-GPI interferes with Von Willebrand factor–dependent platelet activation; neutralization of this effect by β2-GPI antibodies may also potentiate the risk of thrombosis. aPL also induces platelet activation, leading to increased platelet adhesion and thromboxane synthesis. In addition, aPL may confer acquired resistance to the anticoagulant properties of protein C and annexin A5, further contributing to the increased risk of thrombosis. aPL displaces annexin A5 from the trophoblastic surface, which may contribute to pregnancy loss. Important to note, aPL-mediated thrombosis and fetal loss are complement-dependent, and heparin therapy acts via

complement inhibition rather than as an anticoagulant as previously believed.

Clinical Features A subset of people with aPL antibodies eventually exhibit manifestations of APS. Both arterial and venous thrombosis may occur in vessels of any size. Venous thrombosis, especially of the lower extremities, is by far the most common manifestation. Up to half of affected patients develop associated pulmonary embolism. The most common site of arterial thrombosis is the central nervous system, with manifestations including seizures, stroke, dementia, and psychosis. Transverse myelitis, chorea, and Guillain-Barré syndrome have also been reported. In addition, microthrombotic disease is being recognized as a manifestation of APS. Renal failure may occur as a result of microthrombosis and renal artery thrombosis. Cardiac valve thickening (Libman-Sacks endocarditis) is frequently found in patients with APS. These noninfectious vegetations may be a source of embolism. Any valve can be affected, with mitral regurgitation being the most frequent hemodynamic abnormality, followed by aortic regurgitation. This is frequently asymptomatic; and the need for surgery is uncommon. Unlike rheumatic heart disease and bacterial endocarditis, valvular thickening is diffuse and rarely causes rupture. A major feature of APS is recurrent pregnancy loss, especially after the 10th week of gestation. Fetal distress, intrauterine growth retardation, premature births, uteroplacental insufficiency, and oligohydramnios contribute to major pregnancy morbidity. Preeclampsia and HELLP syndrome (hemolysis, elevated liver enzymes, and low platelet count) are also frequent complications. Other prominent manifestations of APS include mild thrombocytopenia, with platelet count usually above 50,000/ mL; APS seldom causes hemorrhage. Hemolytic anemia is relatively uncommon and may occur simultaneously with thrombocytopenia (E

Cecil Essentials Of Medicine - 8th Edition - PDFCOFFEE.COM (2025)